Interpreting the Australian income tax definition of ... · This article takes the High Court’s rejection of ritual incantation seriously by critically analysing the elements of

eJournal of Tax Research (2018) vol. 16, no. 1, pp. 2-36

2

Interpreting the Australian income tax

definition of ‘ordinary income’: ritual

incantation or analysis, when examined through

the lens of early twentieth century linguistic

philosophy?

Mark Burton

Abstract

This article examines the interpretation of subsection 6-5(1) of the Income Tax Assessment Act 1997 (Cth). The High Court of

Australia has concluded that subsection 6-5(1) should be understood upon the basis that it adopts some or all of the elements

of the oft-quoted dictum of Sir Frederick Jordan in Scott v Commissioner of Taxation. The article shows that different decisions

of the High Court of Australia have adopted different statements of the elements of the dictum that ought to be referred to, and

some decisions of the High Court have ignored the dictum altogether. Moreover, no High Court decision has analysed the

statement of Jordan CJ. This article addresses this absence of analysis by offering an analysis of that dictum having regard to

the vibrant, contemporaneous linguistic philosophy of the early twentieth century. This analysis suggests that Sir Frederick

Jordan’s dictum is vague and appears to incorporate unresolved contradictions.

Key words: income tax law, ordinary income, statutory interpretation, linguistic philosophy

Associate Professor, Melbourne Law School.

eJournal of Tax Research Interpreting the Australian income tax definition of ‘ordinary income’

3

1. INTRODUCTION

At the heart of the Australian income tax one of the core assessing provisions – sub-

section 6-5(1) of the Income Tax Assessment Act 1997 (Cth) – states that ‘your

assessable income includes income according to ordinary concepts, which is called

ordinary income’. This ‘plain language’ definition was adopted in the context of

common public acknowledgement that a clearer definition of ‘income’ was needed.1

However, statutory guidance regarding the meaning of ordinary income is extremely

limited.2 Likewise, the extrinsic materials provide no clear guidance upon the meaning

of the statutory terms.3

In the course of considering subsection 6-5(1), the High Court of Australia has

emphasised the significance of the statement of Jordan CJ in Scott4 and has indicated

that several elements of his Honour’s statement ‘are no mere matters of ritual

incantation; they identify the essential nature of the inquiry’.5 Somewhat confusingly,

different majority decisions in the High Court have adopted different statements of the

elements to be extracted from Sir Frederick Jordan’s statement6 or have even ignored

this ‘essential nature of the inquiry’ altogether.7 This confusion is also manifest in the

specification of the test in the lower courts,8 in extrajudicial discussion9 and in

extrajudicial scepticism as to the basis upon which amounts are characterised for income

1 Joint Committee of Public Accounts, Commonwealth Parliament, An Assessment of Tax, Report No 326

(1993) 76. Prior to the enactment of the Income Tax Assessment Act 1997 (Cth), Australian income tax

legislation at both federal and state levels referred to the general concept of ‘income’ without

comprehensively defining that term for the purposes of the income tax legislation. The Income Tax

Assessment Act 1915 (Cth) referred to income, providing an inclusive definition in section 3 which stated

that ‘“Income” includes interest upon money secured by mortgage of any property in Australia’. Section 3

of the Income Tax Assessment Act 1922 (Cth) expanded this inclusive definition by including within the

concept of ‘income’ some payments to members of cooperatives while excluding rebates paid to members

of cooperatives as well as the component of a purchased annuity that is attributable to the purchase price of

that annuity. Section 25 of the Income Tax Assessment Act 1936 (Cth) referred to ‘income’. For the purpose

of differentiating between different classes of income, all of these earlier Acts regarding income tax defined

‘income from personal exertion’ and ‘income from property’. However, as noted in many case decisions,

these definitions included ‘income’ on both sides of the definition and therefore did not define income

itself. 2 Note the definitions of ‘income from personal exertion’ and ‘income from property’, which generally

have been regarded as providing little guidance regarding the concept of income because they refer to

income on both sides of the definition: per Windeyer J in Scott v Federal Commissioner of Taxation (1966)

117 CLR 514, 524 (citing Jordan CJ in Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 219);

note, however, the apparent reliance upon the definition of income from personal exertion in Stone, without

reference to the earlier case law and without close analysis of the limitations of the statutory definition:

Commissioner of Taxation v Stone (2005) 222 CLR 289, 296-297 [17]. 3 Chief Justice Robert French, ‘Law – Complexity and Moral Clarity’ (2013) 40(6) Brief 25, 25; the limited

usefulness of the Explanatory Memorandum is critically assessed in a second article under preparation by

this author. 4 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 220. 5 Federal Commissioner of Taxation v Montgomery (1999) 198 CLR 639, 661 [64]. 6 Discussed in section 2 of this article. 7 Federal Commissioner of Taxation v McNeil (2007) 229 CLR 656. 8 Federal Commissioner of Taxation v McNeil [2005] FCAFC 147, 144 FCR 514, per French J, [40]; cf the

fuller statement of Sir Frederick’s approach by Dowsett J in McNeil, [2005] FCAFC 147 [207]. 9 G T Pagone, ‘Income and Capital Distinction’, paper delivered at the South Australian Convention of the

Taxation Institution of Australia, 9 May 2009, 2, citing Scott as authority for the process for determining

income specified by Dixon J in The Commissioner of Taxes (South Australia) v The Executor Trustee and

Agency Company of South Australia Limited (Carden’s Case) (1938) 63 CLR 108, 152.


4

tax purposes.10 Writing extrajudicially, Chief Justice French opined that judicial

formulations of the income concept were ‘broad, but almost content-free,

generalisations’.11 Referring to the High Court decision in McNeil,12 his Honour

observed that the decision illustrates how ‘[a] simple broadly expressed provision on

the one hand attracts a plethora of judicial exposition such that some would say the true

meaning of the statute, if it has one, is buried in the cases’.13

This confusion regarding the terms that subsection 6-5(1) is taken to incorporate, and

also the meaning of those terms, is difficult to dispel because of the absence of close

analysis of the elements of Sir Frederick Jordan’s statement in the High Court decisions

that apparently adopt that statement.14 The absence of such analysis is cause for

questioning the High Court’s observation that reference to the Jordan CJ definition is

not a matter of ritual incantation.15 If Sir Frederick Jordan’s statement were to be

analysed, what would this analysis reveal? Would it reveal what might hitherto have

been the hidden elements of the concept of income according to ordinary concepts? Or

would such analysis expose irresolvable tensions between the elements of that

statement?

This article takes the High Court’s rejection of ritual incantation seriously by critically

analysing the elements of Sir Frederick Jordan’s statement. The purpose of this analysis

is to determine whether that statement provides secure foundations for the concept of

income according to ordinary concepts. The reference to ‘the essential nature of the

inquiry’16 in Montgomery implies that firm intellectual footings can be found in Sir

Frederick Jordan’s statement. The narrative here is of an objectively specified inquiry

that constitutes the foundation for identification of the ‘ordinary income’ component of

the Australian income tax base.17 This narrative simultaneously draws upon and

reinforces the ‘orthodox’ narratives of the rule of law and judicial constraint, which

10 ‘I suspect, but can not verify, that most judges decide first the outcome they wish to reach and then use

the appropriate maxim of interpretation to justify it, rather than first applying the maxim of interpretation

to reach the outcome’: Justice D Graham Hill, ‘A Judicial Perspective on Tax Law Reform’ (1998) 72(9)

Australian Law Journal 685, 686. 11 French, above n 3, 25. 12 Federal Commissioner of Taxation v McNeil (2007) 229 CLR 656. 13 French, above n 3, 28. 14 In another paper my review of relevant High Court decisions indicates that no such analysis is to be found

in the respective majority judgments in those case decisions. 15 See, for example, Foots v Southern Cross Mine Management Pty Ltd (2007) 234 CLR 52, 73 [55]: ‘British

Gold Fields is a case whose authority stems more from repetition than from analysis’, per Gleeson CJ,

Gummow, Hayne and Crennan JJ. It is possible that this close analysis has been undertaken but not written

into the relevant High Court judgments, in which case those case decisions do not transparently reflect all

of the significant logical steps in reasoning to the conclusions reached. See, for example: Sir Frank Kitto,

‘Why Write Judgments?’ (1992) 66 Australian Law Journal 787; Sir Harry Gibbs, ‘Judgment Writing’

(1993) 67 Australian Law Journal 494; Stephen Gageler, ‘Why Write Judgments?’ (2014) 36(2) Sydney

Law Review 189. 16 Federal Commissioner of Taxation v Montgomery (1999) 198 CLR 639, 661 [64]. 17 See, for example, Matthew Kramer, Objectivity and the Rule of Law (Cambridge University Press, 2009)

ch 2; Larry Alexander, ‘The Objectivity of Morality, Rules, and Law: A Conceptual Map’ (2013) 65(2)

Alabama Law Review 501; Nicos Stavropoulos, Objectivity in Law (Oxford University Press, 1996). For

the view that ‘legislative intention’ is the relevant ‘object’, see Richard Ekins and Jeffrey Goldsworthy,

‘The Reality and Indispensability of Legislative Intentions’ (2014) 36(1) Sydney Law Review 39.


5

depict judges applying more or less determinate law in a manner consistent with the rule

of law.18

My argument in this article is that judicial analysis of Sir Frederick Jordan’s judgment

is necessary if the High Court is to move beyond ritual incantation of the statement that

apparently specifies the essential nature of the inquiry into the meaning of income.

Further, I suggest that such analysis is best undertaken through the lens of the vibrant

debate regarding linguistic philosophy which emerged as Western philosophy worked

itself through its ‘linguistic turn’ from the late nineteenth century.19 That debate

included close consideration of different accounts of linguistic meaning. On one view

advanced in the course of this debate, any linguistic element (a word, phrase, sentence,

etc) embodied the formal specification of its meaning. Another view was that the

meaning of any linguistic element was to be found in the conventional usage of that

element. To be clear, this article is not presenting the argument that the dictum of Jordan

CJ in Scott is directly or indirectly referring to this philosophical literature. The

judgments of Jordan CJ do not expressly refer to this philosophical literature, and nor

does biographical material expressly state that Jordan CJ was aware of this literature

notwithstanding his keen interest in literature.20 However, the reason for referring to this

philosophical literature is that it expresses most clearly the different views regarding

sources of linguistic meaning that are commonly referred to by us all when we examine

the nature of linguistic meaning. When we discuss linguistic meaning we commonly

refer to these different sources of meaning. We might say ‘we use “income” to mean

…’ and we might also say ‘the semantic meaning is …’. The semantic and pragmatic

strands of linguistic philosophy are not distinctly ‘philosophical’ – they echo common

understandings of competing foundations of linguistic meaning.

When viewed through this lens, it can be seen that Sir Frederick Jordan’s statement

gains rhetorical force from the apparent synthesis of the intellectual foundations that

underpinned different strands of early twentieth century linguistic philosophy and also

common understandings of the foundations of linguistic meaning. I argue that this

rhetorical force should be tested by critically assessing whether Sir Frederick

successfully threaded disparate strands into a durable fabric that underpins a stable and

determinate meaning of ‘income’ or ‘ordinary income’.21 In undertaking this analysis, I

argue that the elements of Sir Frederick’s statement are ill-defined and quite possibly

reflect fundamentally irreconcilable approaches to specifying the nature of the inquiry

into income and also the nature of income itself. The significant practical and theoretical

18 French, above n 3, 30. See also Murray Gleeson, ‘The Meaning of Legislation: Context, Purpose and

Respect for Fundamental Rights’ (2009) 20(1) Public Law Review 26, 27. 19 For a useful critical overview of the development of linguistic philosophy over the latter part of the

nineteenth century and the first half of the twentieth century see P M S Hacker, ‘The Linguistic Turn in

Analytic Philosophy’ in Michael Beaney (ed), The Oxford Handbook of the History of Analytic Philosophy

(Oxford University Press, 2013) 926. 20 See J M Bennett, Portraits of the Chief Justices of New South Wales 1824-1977 (John Ferguson, 1977);

J M Bennett, A History of the Supreme Court of New South Wales (Law Book Co, 1974); Maurice Byers,

‘Recollections of Sir Frederick Jordan’ (1991)(Winter) Bar News 13. This material notes that Sir Frederick

Jordan had a keen interest in literature, but does not directly refer to any awareness of the philosophy of

language. 21 Hacker, above n 19, noting that analytical school may still have several decades of life. With respect to

tax jurisprudence, there is ongoing interest in Wittgenstein’s work, which was a key legacy of the linguistic

turn: Bret N Bogenschneider, ‘Wittgenstein on Why Tax Law is Comprehensible’ [2015] 2 British Tax

Review 252.


6

consequences of these conclusions are noted briefly in the conclusion of this article and

chart a future research program.

2. SIR FREDERICK JORDAN’S STATEMENT IN SCOTT

Sir Frederick Jordan’s statement was made in relation to the former New South Wales

income tax legislation.22 That legislation imposed tax upon ‘taxable income’23 which

was defined to be the amount of assessable income left after taking away allowable

deductions.24 Assessable income was defined to be the gross income after excluding

amounts of income specifically exempted.25 In effect, the concept of income was left

undefined because ‘income’ was defined to mean ‘income derived … directly or

indirectly …’.26 In Scott the taxpayer was appointed to a statutory Board and that Board

was subsequently dissolved by the Meat Industry (Amendment) Act 1932. That Act also

stated that members of the Board should be compensated in the amount that they would

have received had their services been terminated otherwise than according to law. The

taxpayer received a lump sum of £7,000 as compensation. By way of case stated, the

New South Wales Court of Appeal was asked to determine whether the compensation

received by Mr Scott was ‘income’.

In this context, and in deciding that the lump sum was not ‘income’, Jordan CJ27

observed that the assessment of ‘income’ appeared in the context of income tax

legislation, and continued:

The word ‘Income’ is not a term of art and what forms of receipts are

comprehended within it, and what principles are to be applied to ascertain how

much of those receipts ought to be treated as income, must be determined in

accordance with the ordinary concepts and usages of mankind, except in so far

as the statute states or indicates an intention that receipts which are not income

in ordinary parlance are to be treated as income, or that special rules are to [be

applied].28

This dictum addresses two fundamental matters: the subject of the inquiry into the

ordinary meaning of a statutory term and also the process by which the ordinary

meaning is extracted from examination of that subject matter. Different interpretations

of the dictum identify differing specifications of the subject matter and the process.

On its face, the first part of the dictum describes a four-stage inquiry, the logical order29

of which is:

1. determine whether the statutory language is a ‘term of art’;

22 Income Tax (Management) Act 1928 (NSW). 23 Income Tax (Management) Act 1928 s 8(1). 24 Income Tax (Management) Act 1928 s 4. 25 Income Tax (Management) Act 1928 s 4. 26 In relation to the definitions of ‘income from personal exertion’ and ‘income from property’, see

comments at n 2, above. 27 Stephen and Street JJ agreeing on this point. Stephen J decided that the amount was assessable as a

‘retiring allowance’. 28 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 219. 29 As opposed to the order in which the stages of the inquiry are identified in the dictum of Jordan CJ.


7

2. if the statutory language is not a ‘term of art’, such as the statutory reference to

‘income’, identify the set comprising ‘ordinary concepts and usages’;

3. review this broad set in order to identify receipts of a form that are

comprehended as income under ordinary concepts and usages; and

4. consider the principles that are applied for determining how much of a receipt

that fits a requisite form ‘ought to be treated as income’.

Thus, for statutory terms that are not ‘terms of art’, the subject matter comprising

‘ordinary concepts and usages’ is analysed for the purpose of identifying the forms and

principles governing inclusion in the relevant set. These forms and principles are then

applied to identify members of the set identified by the statute (such as ‘income’). This

approach examines what might be described as the logical fabric of our language that

may lie ‘beneath’ or be seen to underpin the meaning of terms according to ordinary

concepts and usages.

However, on a different interpretation, the proviso in the latter part of Sir Frederick

Jordan’s statement suggests that different subject matter and a different process should

be adopted. This proviso refers to identifying amounts that are income ‘in ordinary

parlance’. On one interpretation of this approach, a judge would examine ‘ordinary

parlance’ to produce a list of receipts that have come to be labelled as ‘income’ ‘in

ordinary parlance’. On this interpretation, and in contrast to the analytical approach

identified in the first part of the dictum, the second part of the statement focuses upon

the ‘surface meaning’ that is presumably obvious to any observer of ordinary parlance

prepared to compile a list of things referred to by each linguistic ‘label’.30 This

interpretation appears to be supported by Sir Frederick Jordan’s subsequent

observations, arrived at without any preceding express analysis of ‘income’, that Mr

Scott’s compensation sum was not ‘income’ according to its ‘ordinary’31 and/or

according to its ‘natural’32 meaning. An observer could reasonably interpret these

observations to imply that ‘income’ is a label that need not be analysed in order to

identify the set of rules governing inclusion in the set ‘income’ because the term

‘income’ may have come to be applied to all sorts of things by convention. On this

reading of the judgment, we do not need to analyse the meaning of ‘income’ to arrive

at the conclusion that ‘income’ does not refer to a compensation sum such as Mr Scott’s,

because we intuitively know that the label would never be applied to such a sum ‘in

ordinary parlance’.

A third interpretation of Sir Frederick’s reference to ‘ordinary parlance’ could be that

this is no more than a shorthand reference to the analytical extraction of forms/principles

described in the first part of the statement. This approach would overcome the

possibility of any inconsistency between the first and second parts of the statement, but

only by adopting an interpretation of ‘ordinary parlance’ that appears to contradict the

majority’s decision in Montgomery, in which the majority appeared to treat the two

elements as though they were not substitutable.

These tensions within Sir Frederick’s statement will be examined more closely in

section 5 of this article. However, for present purposes suffice it to say that Jordan CJ

30 See the discussion of the naming theory of meaning in section 4.2. 31 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 219. 32 Ibid.


8

might be taken to have contemplated two quite different approaches to identifying the

meaning of ‘income’. Before turning to further analysis of this statement, it is

appropriate to review the adoption of the statement of Jordan CJ by the High Court of

Australia in some decisions upon the meaning of ‘ordinary income’ over the past two

decades.33

3. THE HIGH COURT’S ADOPTION OF THE JORDAN CJ DEFINITION OF ‘INCOME’ WHEN

INTERPRETING SUBSECTION 6-5(1)

In the course of considering the definition of ordinary income in subsection 6-5(1), in

some decisions the High Court has emphasised the significance of the statement of

Jordan CJ. The purpose of this section of the article is to establish two propositions with

respect to the adoption of Sir Frederick’s statement in various High Court decisions.

The first proposition is that the elements of Sir Frederick’s statement have not been

analysed by the High Court. Following from this first proposition, the second

proposition is that this absence of analysis has allowed different judicial paraphrases of

Sir Frederick’s statement to be adopted without criticism. This confusion about the

substantive content of Sir Frederick’s statement is only compounded by the fact that

some High Court decisions dealing with the meaning of ordinary income have not

referred to Sir Frederick’s statement at all, leaving open the possibility that Sir

Frederick’s statement may not be as significant as other High Court decisions suggest.34

In dealing with the application of the former section 25(1) of the Income Tax Assessment

Act 1936 (ITAA 1936), the majority’s joint judgment in Montgomery observed that

counsels’ submissions had been framed upon analogies to decided cases, and continued:

That approach is often helpful, but resort to analogy should not be permitted to

obscure the essential nature of the inquiry which is to determine whether ‘in

ordinary parlance’ the receipt in question is to be treated as income. As Jordan

CJ made plain, the references to ‘ordinary parlance’ and to the ‘ordinary

concepts and usages of mankind’ are no mere matters of ritual incantation; they

identify the essential nature of the inquiry.35

33 For a review of Australian decisions in which the dictum of Jordan CJ was directly referred to, see Mark

Burton, ‘A review of judicial references to the dictum of Jordan CJ, expressed in Scott v Commissioner of

Taxation, in elaborating the meaning of “income” for the purposes of the Australian income tax’ (2017) 9

Journal of Australian Taxation 297. 34 This article clearly proceeds upon the basis that judicial words within a judgment have a meaning and

that this meaning is important to the legitimacy of judgments. The legitimacy of judgments, according to

the widely accepted understanding of law framed in terms of ‘legal objectivity’, exists because a judgment

explains how the outcome of the case was arrived at ‘according to law’. The law is an object which pre-

exists any judgment and so the judge discovers the relevant law and applies that law to the facts of the

particular case. Taking judicial words in judgments seriously, by analysing those words in pursuit of an

understanding of the judicial discovery of the statutory meaning, is routine in legal analysis.

Acknowledging this significance of judicial words in judgments is consistent with the principle that judicial

words in a judgment do not supplant the meaning of statutory terms, but rather reveal how the meaning of

statutory terms was discovered and how that meaning was applied in a particular case. In this section of the

article, I discuss the different judicial descriptions of the meaning of the dictum of Jordan CJ that have been

adopted in judgments of the High Court. In doing so, I identify differing verbal formulations that appear to

describe the meaning of that dictum. I finish this section by noting that this proliferation of judicial

formulations of the meaning of Sir Frederick Jordan’s dictum is problematic if we accept the proposition

that the statutory text in subsection 6-5(1) only has one meaning which, according to the widely accepted

account of legal objectivity, pre-exists any case decision. 35 Federal Commissioner of Taxation v Montgomery (1999) 198 CLR 639, 661 [64].


9

This statement of the majority in Montgomery indicates that Sir Frederick’s statement

should be the starting point for any consideration of whether a particular amount is

‘income’. However, according to its terms, this statement differs from that of Jordan CJ

in several important respects.

First, Montgomery does not expressly refer to the forms/principles analysis described

by Jordan CJ.

Second, Montgomery appears to identify several quite different and incomparable

‘essential natures of the inquiry’ in the statement of Jordan CJ. The first such

specification of the relevant inquiry is expressly identified as the identification of

whether ‘in ordinary parlance’ a receipt is ‘to be treated as income’. However, this

statement could be interpreted in different ways. This could be a reference to the

forms/principles analysis of Jordan CJ outlined above. Alternatively, the statement

could be a reference to the ‘surface meaning’ approach described by Jordan CJ in the

second part of his Honour’s statement, also outlined above. As a further possible

alternative, could ‘is to be treated as income’ be taken to mean that the courts will

determine whether a particular amount ‘is to be treated as income’ and that this

stipulation of meaning thereafter will be adopted in ordinary parlance?

The second essential nature of the relevant inquiry identified in Montgomery requires

reference to ‘ordinary parlance’ and to ‘the ordinary concepts and usages of mankind’.

However, although the majority in Montgomery stated that these phrases describe the

nature of the inquiry, these phrases do not describe a process of inquiry but do

potentially identify subjects of an inquiry. The phrases identify what is to be examined

without specifying the process of examination. Moreover, the two phrases are not

synonymous. ‘Ordinary concepts and usages’ could include written usages, for example,

and ‘parlance’ generally would not refer to written usage.36 Given this potential conflict,

the relationship of these two elements should have been made clear in Scott and/or in

Montgomery. In the absence of any judicial elaboration of these potentially inconsistent

elements, the application of Sir Frederick’s statement remains problematic after

Montgomery.

In Federal Commissioner of Taxation v Stone37 the High Court was called upon to

determine whether receipts of a sportsperson were ‘ordinary income’ for the purposes

of section 6-5 of the Income Tax Assessment Act 1997 (ITAA 1997). The ‘plain

language’ definition of ‘ordinary income’ in subsection 6-5(1) was quoted at the

beginning of this article. One question that arose was whether this definition

incorporated a change to the law. The former legislation merely referred to ‘income’

while the new legislation defined the new statutory concept of ‘ordinary income’ to be

‘income according to ordinary concepts’. There was no comparable ‘old law’ definition

of income and the definition of ‘ordinary income’ clearly did not duplicate all of Sir

Frederick’s statement.

In the course of deciding that Stone’s receipts were ordinary income, the joint judgment

referred to ITAA 1997, section 1-338 and immediately concluded that subsection 6-5(1)

36 See the discussion of this matter in section 4.8 below. 37 (2005) 222 CLR 289. 38 The High Court did not analyse the text of section 1-3.


10

was an ‘evident reference’39 to the decision of Jordan CJ in Scott. Sir Frederick’s

statement in Scott was then extracted in full.

Later in the joint judgment in Stone, the plurality noted that the existence of a business

activity ‘perhaps very often’ will carry with it the conclusion that the proceeds of that

business activity are ordinary income.40 However, the joint judgment appears to have

adopted a definition of ordinary income that is not necessarily supported by the text of

subsection 6-5(1) and nor by the statement of Sir Frederick Jordan:

Asking whether a person was carrying on a business may therefore be useful

and necessary. But the inquiry about “business” must not be permitted to

distract attention from the question presented by both the 1936 Act and the 1997

Act. That question seeks to identify whether a receipt is, or receipts are,

“income”. As s 6-5 of the 1997 Act makes plain, that requires consideration of

whether the receipt in question is income in accordance with “the ordinary

concepts and usages of mankind” (emphasis added).41

With respect, the adoption of Sir Frederick’s statement in Stone is open to criticism.

One cause for criticism is that the text of subsection 6-5(1) is substantially different to

the text of Sir Frederick’s statement. Given this difference, the conclusion that

subsection 6-5(1) adopted all of Sir Frederick’s statement does not necessarily follow

from the premise that subsection 6-5(1) incorporates one phrase from Sir Frederick’s

statement. If anything, there is a good case for concluding that the more narrowly framed

statutory text should prevail over the inquiry more fully expressed by Sir Frederick,42

rather than taking the statutory text to have adopted the entire dictum of Jordan CJ.

Moreover, the joint judgment in Stone did not undertake any express textual analysis of

ITAA 1997, section 1-3, but stated that ‘[b]ecause the 1997 Act contains provisions of

the 1936 Act in a rewritten form, construing the word “income” in the 1997 Act requires

reference to the definition in s 6(1) of the 1936 Act of “income from personal

exertion”’.43 This statement indicates that the majority accepted that the provisions of

the ITAA 1997 necessarily had the same meaning as the comparable provisions found

in the ITAA 1936. However, this conclusion is inconsistent with section 1-3 itself,44 the

statutory framework of the ITAA 199745 and the relevant statement in the Explanatory

39 Commissioner of Taxation v Stone (2005) 222 CLR 289, 294 [8]. 40 Ibid 296 [16]. 41 Ibid. 42 See, for example, FCT v Slater Holdings Limited (1954) 15 CLR 447; Gageler J in Baini v The Queen

[2012] HCA 59; 24 CLR 469 [43]. 43 Commissioner of Taxation v Stone (2005) 222 CLR 289, 296-297 [17]. This statement appears to be a

reference to the earlier statement that ‘section 1-3(1) of the 1997 Act provides that the 1997 Act contains

provisions of the 1936 Act ‘in a rewritten form’: ibid [9]. 44 Beginning with the conditional conjunction ‘if’, subsection 1-3(2) requires consideration of whether it is

true that the ITAA 1997 appears to have expressed the same idea as that which was expressed in the ITAA

1936. On the reading of subsection 1-3(1) adopted in Stone, the test set out in subsection 1-3(2) would

necessarily be answered in the affirmative. In this case, why would the legislature have used the conditional

‘if’? 45 Section 15-2, for example, clearly recognises that it is subordinate to the general income rule in section

6-5(1). By contrast, the former comparable rule (ITAA 1936 s 26(e)) took priority over the former general

assessing rule in ITAA 1936 s 25(1).


11

Memorandum accompanying the Bill for the ITAA 1997.46 Deciding that section 1-3

applies without analysing its terms first is contrary to the description of orthodox

interpretative practice set out in various High Court decisions.47 This is particularly

significant given the rebuttable presumptive rule of statutory interpretation that different

statutory words are presumed to be intended to have a different meaning.48

The joint judgment in Stone refers to the decision in Montgomery in support of other

propositions regarding the concept of ordinary income.49 However, the joint judgment

in Stone does not directly refer to the consideration of Scott in Montgomery. This

oversight is extraordinary, given the observation in Montgomery that the decision in

Scott specifies the essential nature of the relevant inquiry into the nature of income.

Finally, at different points the joint judgment appears to adopt different elements of Sir

Frederick’s statement. At paragraph 8 the joint judgment extracted Sir Frederick’s

statement in full, which suggests that the entire statement ought to be adopted for the

purpose of identifying ‘income’. At paragraph 16, extracted above, the joint judgment

appears to adopt just one part of Sir Frederick’s statement: ‘income in accordance with

“the ordinary concepts and usages of mankind”’.

In FCT v Anstis50 the joint judgment in the High Court also appears to have adopted an

ambiguous position with respect to Sir Frederick Jordan’s statement by referring to the

decision in Stone with approval, by adopting the limited reference to elements of Sir

Frederick’s statement in Montgomery51 and by paraphrasing that statement with full

reference to the forms/principles inquiry:

As has been said [citing Stone], that is an evident reference to the statement by

Jordan CJ that the forms of receipt falling within the term ‘income’, and the

principles to be applied to ascertain how much of those receipts ought to be

treated as income, ‘must be determined in accordance with the ordinary

concepts and usages of mankind, except in so far as the statute states or

indicates an intention that receipts which are not income in ordinary parlance

are to be treated as income’ [citing Scott]. The reference to ‘ordinary parlance’

and to the ‘ordinary usages of mankind’ are ‘no mere matters of ritual

incantation; they identify the essential nature of the inquiry’ [citing

Montgomery].52

In Commissioner of Taxation v McNeil53 the joint judgment made no direct reference to

Sir Frederick Jordan’s statement or to the High Court decisions adopting that statement

for the purposes of subsection 6-5(1). Rather, the decision in McNeil was grounded upon

46 The Explanatory Memorandum noted that ‘most provisions are being rewritten without any intention of

changing their effect’: Explanatory Memorandum accompanying the Income Tax Assessment Bill 1996,

34. 47 Project Blue Sky Inc v Australia Broadcasting Authority (1998) 194 CLR 355, 381-382 [69]. 48 D C Pearce and R A Geddes, Statutory Interpretation in Australia (LexisNexis Australia, 8th ed, 2014)

150-154 [4.6]-[4.7]. 49 See Commissioner of Taxation v Stone (2005) 222 CLR 289, 297 [18] and 306 [60]. 50 (2010) 241 CLR 443. 51 Federal Commissioner of Taxation v Montgomery (1999) 198 CLR 639, 661 [64]. 52 Federal Commissioner of Taxation v Anstis (2010) 241 CLR 443, 450 [13]. 53 Commissioner of Taxation v McNeil (2007) 229 CLR 656.


12

acceptance of Justice Pitney’s statement of principle in Eisner v Macomber,54 as adopted

by the High Court in Montgomery.55

In Spriggs the unanimous Full High Court decision observed:

It was not disputed that these payments were income according to the concept

of ‘ordinary income’ under s 6-5 of the ITAA, that is, within ‘the ordinary

concepts and usages of mankind’ [citing Scott].56

The endnote to this text referred to the decision of Jordan CJ in Scott, the decision in

Stone and also to the Explanatory Memorandum accompanying the Income Tax

Assessment Bill 1996.

In section 2 of this article I suggested that Sir Frederick Jordan’s statement in Scott

incorporated two apparently different approaches to identifying ‘income’ and,

moreover, that his Honour did not elaborate upon the elements of these approaches in

the course of his judgment. The review of High Court judgments dealing with Sir

Frederick Jordan’s statement in this section of the article demonstrates that those

decisions have not incorporated express, close analysis of that statement. This absence

of close analysis of Sir Frederick Jordan’s definition is striking, given the reasons for

the preparation of written judgments in the appellate courts.57 Arguably, express

analysis of Sir Frederick’s judgment would have shone light upon the meaning and

application of the elements of Sir Frederick’s statement. This critical appraisal of Scott

would have avoided the omissions and inconsistencies regarding this matter that can be

seen in the decisions in Montgomery, Stone, Anstis, McNeil and Spriggs. Upon the basis

of those decisions, there are at least five, and potentially six, different approaches to the

relevance of Sir Frederick’s statement for the purpose of identifying ‘ordinary income’:

1. the full statement of Jordan CJ in Scott (Stone, Anstis);

2. ‘income according to the ordinary concepts and usages of mankind’ (Stone,

Spriggs);

3. ‘ordinary parlance’ (Montgomery);

4. ‘ordinary concepts and usages of mankind’ and ‘ordinary parlance’

(Montgomery);

5. the statement may not be relevant to the determination of some ‘income’ cases,

given the adoption of a different approach in McNeil with respect to income

from property without any reconciliation to Sir Frederick’s statement; and

6. apply the terms of the statutory text – ‘income according to ordinary concepts’

– given that the general rules of statutory interpretation place considerable

emphasis upon the statutory text, particularly in the absence of any clear

54 252 US 189, 206-207 (1920). 55 (1999) 198 CLR 639, 660-663. See critical discussion of these judgments in Neil Young QC, ‘The

Historical Significance of the High Court’s Decision in Federal Commissioner of Taxation v The Myer

Emporium Ltd’ (2007) 31(1) Melbourne University Law Review 266, 285ff. 56 Spriggs v Federal Commissioner of Taxation (2009) 239 CLR 1, 17. 57 See the material referred to at n 15 above.


13

rationale for displacing the statutory text of subsection 6-5(1) with some other

judicial words more or less drawn from the statement of Sir Frederick in Scott.

This proliferation of general statements of principle regarding the process for

identifying the meaning of ‘ordinary income’, apparent within particular case decisions

(eg, Stone) as well as across different decisions, should be of concern to any observer

who considers that the law ‘works itself pure’ by progressively identifying the

respective determinate meanings of statutory terms. These general statements are not

tailored to the facts of particular cases because they describe the process for identifying

the meaning of ordinary income, rather than the outcome of the process, the meaning of

income. The process for identifying the meaning of ordinary income should be the same

in any case, and this is reflected in the singular reference to ‘the essential nature of the

inquiry’ in Montgomery.

4. IDEAS OF ‘MEANING’ AROUND THE TIME OF SIR FREDERICK JORDAN’S STATEMENT

In the absence of judicial analysis of Sir Frederick’s statement, different commentators

could adopt different approaches to elaborating the meaning and significance of that

statement. I have decided to examine the statement by locating it within its historical

context, and in particular by critically analysing that statement through the lens of

linguistic philosophy that so preoccupied early twentieth century Western philosophers.

By doing so I am not implying that judges are or should be philosophers.58 However, as

will become apparent, analysis of the meaning of the elements of Sir Frederick Jordan’s

statement in Scott is enhanced by examining those elements in the context of the themes

within the contemporaneous linguistic philosophy.

Two matters are central to identifying the meaning of legislation:

1. In general terms, how does language ‘latch onto’ ideas and things in our world

so that we can use language to communicate about those ideas and things?

For example, when we use the word ‘income’ to describe a particular amount,

what is it that we are referring to? Does the ‘incomeness’ of the amount inhere

in the essential nature of the particular amount, in its DNA if you like? Or is the

incomeness of the amount determined irrespective of the essential nature of the

amount itself, and found rather in the semantic or conventional rules governing

our use of the label ‘income’ when we perceive and understand our world? Or

is ‘income’ whatever the legislature intended should be treated as income?

Our answers to these questions determine what we examine when identifying

the meaning of statutory terms – the intrinsic nature of the things in the world

to which we attach our linguistic labels, our linguistic conventions regarding

the semantic meaning of linguistic symbols or the rules of usage regarding those

symbols or the intention of the ‘speaker’ who uttered the linguistic symbols.

58 This matter has attracted opposition from legal positivists such as Joseph Raz, apparently upon the basis

that philosopher judges would import metaphysics into the law - a proposition anathema to legal positivism:

Joseph Raz, Between Authority and Interpretation (Oxford University Press, 2009) 53, 79ff. By contrast,

Ronald Dworkin’s natural law theory accommodated the concept of the philosopher judge who would

grapple with metaphysical principles underpinning the law: Ronald Dworkin, Law’s Empire (Belknap

Press, 1986) 90.


14

This is the metaphysical aspect of statutory interpretation; and

2. Once the first question above is answered, what is the process by which we

identify the meaning of the language unit and how do we specify the standard

to be applied in determining that a claim to have found that meaning should be

accepted as true?

This is the epistemological aspect of statutory interpretation.59

This brief reference to the interrelated metaphysical and epistemological questions is

sufficient to indicate that a meaningful definition of ordinary income will explicitly

and/or implicitly answer both questions by telling us what makes something ‘income’

and also by telling us how to go about verifying the truthfulness of a statement that a

particular amount is/is not income. Some or even all of these matters might be addressed

by expressly or implicitly adopting social norms, such as the norms of logic.60

4.1 What is statutory language referring to? The concept of income and the problem of

universals

The concepts of ‘ordinary’, ‘usage’, ‘parlance’, ‘concept’ and ‘income’ are all instances

of universals, in that they name general categories of particular instances.61 When we

examine a particular monetary sum (say, ‘the $1,000 paid by Tom to Mary at 8:01am

on 15 June 2016’) to determine whether it falls within the set of monetary sums that we

call ‘income’, we are confronting what philosophers refer to as the problem of

universals. The problem of universals grapples with the basis upon which we determine

that a particular instance is appropriately determined to fall within a universal category.

Commentators have identified three broad categories of theories regarding the problem

of universals: realism, conceptualism and nominalism.

4.1.1 Realism

Realist approaches to this metaphysical question share the idea that things that we know

of or perceive exist and also exist independently of our thoughts or beliefs about those

objects of our thought. Some commentators appear to have adopted some form of

realism when considering the basis upon which particular amounts are identified as

income. Prebble, for example, suggests that a particular receipt has a substantive essence

or nature that legal characterisations can only imperfectly identify.62 Upon this basis,

Prebble argues that the law of income taxation must always be incomprehensible

because it can never hope to operate upon the ‘real’ natures of particular amounts.63

One strand of realism, developed by Plato, holds that a universal is an ideal form of a

property that exists independently of any particular instance of the universal and is also

manifest in each particular instance of that universal. The ideal Form exists

independently of thought, time and space, being received by humans prior to birth and

‘remembered’ afterwards by an intellectual elite (philosophers). Reference to a

59 Michael Moore, ‘A Natural Law Theory of Interpretation’ (1985) 58(1-2) Southern California Law

Review 277, 291. 60 Raz, above n 58; Ekins and Goldsworthy, above n 17, 67. 61 Aristotle defined universals as being ‘predicated of many’: Aristotle, ‘De Interpretatione’, (17a38) in

Aristotle, Categories and De Interpretatione (trans J L Ackrill, Clarendon, 1963) 37. 62 John Prebble, ‘Income Taxation: A Structure Built on Sand’ (2002) 24(3) Sydney Law Review 301, 306. 63 John Prebble, ‘Why is Tax Law Incomprehensible?’ [1994] 4 British Tax Review 380, 388.


15

universal such as ‘income’ therefore entails reference to a thing that exists, the Form of

income. In Plato’s metaphor of the cave, universals (Forms) exist behind observers and

the shadows of the Forms are seen on a cave wall. Unable to see the Forms directly, the

observers are consigned to only ever seeing shadows of those Forms. One reading of

the Cave metaphor is that ordinary people live their respective lives without ever seeing

the purity of Forms, which have to be rediscovered by philosophers who reveal this a

priori knowledge by rational thought, rather than discovering knowledge by empirical

investigation of the world.

Another form of realism is attributed to Aristotle. In broad terms, one reading of

Aristotle64 is that he accepted that universals exist, but unlike Plato, Aristotle considered

that universals exist only in the particular forms of matter65 where a universal is

manifest. According to this view, the substance of things is their respective primary

essences and there is a difference between these primary essences and accidental

characteristics.66 Tom’s payment to Mary may be made by way of 10 x $100 currency

notes rather than 20 x $50 currency notes. The mode of payment is ‘accidental’ in that

it does not affect the essential nature of the payment.

After ruling out various contenders for identifying the essence of something upon

grounds that need not detain us here, Aristotle proposed that the essence of something

is its ‘form’ that inheres in the thing rather than the matter of which it is composed. The

essence of a bronze statue is the form of the statue rather than the bronze (ie, the

‘matter’) of which the statue is made. Likewise, the essence of Tom’s payment to Mary

is the form of the payment rather than the (possibly ‘nonsubstantial’) ‘matter’ of it.

According to one reading of Aristotle, the form of something is a universal or a

compound of universals. Universals are capable of definition and definitions may

incorporate universals which can be defined.67 This process of definition continues until

the simple, logical atoms of the definition have been identified.

So on this reading of Aristotle the substance of income is its essence, its essence consists

of its form and its form consists of universals and there are definitions of universals in

terms of necessary and sufficient conditions that must be found in any particular

instance of ‘income’. On this view, when our senses perceive an amount that is

‘income’, the image of income that we receive corresponds to the definition of ‘income’

because the form of the particular image of the amount corresponds to the definition of

‘income’. For example, the manifestation of the form of a thing (comprising of

universals) can be demonstrated by matching the necessary and sufficient criteria of the

universal to the essence of the particular instance (where the essence will assume a

particular form and a form will be or comprise definable universals).

It is important to note that ‘form’ typically differs from what lawyers refer to as the legal

form of a transaction because the legal rules governing the nature of the legal form of a

thing do not necessarily correspond to the universals that realist philosophers refer to.

64 For the contrary argument, that each form of a kind is unique to each particular instance of that kind, see

Edwin Hartman, ‘Aristotle on the Identity of Substance and Essence’ (1976) 85(4) Philosophical Review

545; Charlotte Witt, ‘Aristotelian Essentialism Revisited’ (1989) 27(2) Journal of the History of Philosophy

285. 65 Here, matter can be perceived (ie, the wood in a tree) and intelligible (such as a geometric figure). 66 See, for example, Aristotle, Metaphysics, Book 5 ch 6, observing that a person being musical is taken to

be an accidental characteristic of that person rather than essential to that person’s being. 67 Aristotle, Metaphysics, Book Z8, 1034a6-8.


16

Thus, as I have already noted, Prebble argued that the legal forms of ‘income’ do not

capture the real form of income.68

4.1.2 Conceptualism

The second and third approaches to the problem of universals reject the suggestion that

universals exist independently of human thoughts and beliefs.

The second approach, referred to as conceptualism, maintains that concepts identify

characteristics that are common across a range of particular instances. Thus, according

to conceptualists, there is a concept of ‘income’ that picks out the characteristics that

make a particular instance, such as Tom’s payment to Mary, ‘income’. By assessing the

extent to which a particular amount satisfies those characteristics, we are able to classify

some monetary amounts as particular instances of income. Concepts therefore exist

independently of the particular instances that they describe, but concepts are not things

that exist in the particular instances of the concepts in the way that realists envisage

universals to be. If concepts are things that exist independently of the things in the world

that they categorise, we need to identify the concept of a concept that we are applying

when identifying these metaphysical things called concepts. The problematic concept

of a concept is discussed further in section 5.9 of this article below.

4.1.3 Nominalism

The third approach to the problem of universals, referred to as nominalism, rejects the

existence of universals from particular instances and also rejects the idea that concepts

mediate between the real world and our perception of the real world. A nominalist

approach to words such as ‘income’ is that they merely describe characteristics that exist

and that humans have identified as part of the reference of that linguistic unit – there is

no need to incorporate metaphysical ‘universals’ or ‘concepts’ within a nominalist

account. Nominalism therefore appears to resolve the problem of universals without

reliance upon excess theoretical components (ie, Plato’s forms, Aristotle’s universals,

or concepts). Rather than focusing upon things that are external to the linguistic unit, a

nominalist can find the meaning of a linguistic unit within that linguistic unit itself by

undertaking semantic analysis of that language unit and/or by considering the rules

governing the use of the particular language unit and/or by finding the meaning intended

by the ‘speaker’.

4.2 The philosophy of language – are ordinary usages sufficient to underpin propositions

regarding truths about the world?

At the time that Jordan CJ set out his definition of income, Western philosophy was

working itself through ‘the linguistic turn’ – a reconsideration of the nature of

philosophy which examined whether and how philosophical problems were linguistic

problems.69 In particular, the nature of any connection between the world and the

truthfulness of statements made about that world was subjected to renewed scrutiny

from the late nineteenth century.

68 See text accompanying nn 62-63 above. 69 For a history and analysis of the concept of the linguistic turn see Hacker, above n 19.


17

Prior to this renewed interest in language as a philosophical problem,70 it was widely

accepted that language meant what it referred to, and the use of language to refer to

something was learned by associating language ‘names’ with particular objects/ideas

and/or with particular sense perceptions of those objects or ideas.71

Several aspects of language usage suggested that this referential theory of language did

not provide a comprehensive account of language meaning. For example, we have

words for things that do not exist (eg, unicorn, ghost) and use those words to make

meaningful statements. This non-referring aspect even extends to names for people that

do not exist (‘Santa Claus’). We also can say meaningful things about things that do not

exist (‘It is true that Santa Claus does not exist’). Further, Frege observed that different

modes of referring expression could refer to the same thing but have different

meanings.72 For example, ‘The Morning Star’ and ‘The Evening Star’ both refer to

Venus but have a different meaning and therefore cannot necessarily be substituted. To

illustrate this point, before hearing the statement I might not know that ‘the morning

star is Venus and that Venus is the evening star’ and so that statement would add to my

knowledge. Yet under a referential theory of language my knowledge should not be

enhanced because the statement is merely saying ‘Venus=Venus=Venus’. Frege also

noted that this difference between sense and reference created circumstances in which

the premises of a statement might be true but the conclusion is false. For example, what

if Tom believes that the morning star is Venus but Tom does not know that the evening

star is Venus? It would be true to state ‘Tom believes that the morning star is Venus’

and also true to state ‘Venus is the evening star’. However, it would be false to state

‘Tom believes that the morning star is the evening star’. All of this pointed to an

important difference to our understanding of meaning – the difference between the

‘sense’ – the way that we convey information about a subject – and the subject itself

(the ‘reference’).

This distinction between sense and reference therefore prompted a shift away from

conceiving of language as a referential reflection of reality (a naming of reality) and

precipitated various endeavours in search of a theory of linguistic meaning not grounded

upon reference to objects in the world.73 Two approaches emerged over the early

decades of the twentieth century and were the subject of close discussion by the time

Jordan CJ delivered his judgment in Scott. The first of these two approaches conceived

of meaning in terms of truth claims derived by analysis of any particular statement.

According to this account, to understand the logic of language would be to understand

the logic of the world.74 The second approach focused upon identifying the meaning of

70 This was not a new problem. In his Cratylos Plato observed the apparent arbitrariness of words as signs

for what they signified, and suggested that there must be a connection between the sign and the signified. 71 The proposition that language comprised mere names of ideas in the human mind can be seen in John

Locke, An Essay Concerning Human Understanding (Oxford University Press, 1975 [1690]) 166; Jeremy

Bentham, Of Laws in General (H L A Hart ed, Athlone Press, 1970 [1782]) 82. 72 Gottlob Frege, ‘On Sense and Reference’ (1892) in Peter Geach and Max Black (eds and trs),

Translations from the Philosophical Writings of Gottlob Frege (Blackwell, 2nd ed, 1960) 57, 62. 73 Writing in his later Philosophical Investigations, Wittgenstein captured the general perception of this

picture theory of language as being one too primitive to capture our description of natural language: Ludwig

Wittgenstein, Philosophical Investigations (tr. G E M Anscombe, P M S Hacker and J Schulte, Wiley

Blackwell, 2009 [1953]) 6 [2]. 74 See discussion of this in P M S Hacker, ‘Wittgenstein’s Anthropological and Ethnological Approach’ in

Jesús Padilla Gálvez (ed), Philosophical Anthropology: Wittgenstein’s Perspective (Ontos Verlag, 2010)

15, 15.


18

a statement by examining the social conventions governing the making of that statement

in the particular circumstances in which it was made.

Over this period Bertrand Russell proposed his theory of descriptions, by which he

hoped to reveal the hidden logic of the complex compositions within singular statements

made in natural language.75 Russell’s study of the internal logic of statements promised

a comprehensive and objective treatment of meaning that focused upon the truthfulness

of the propositions that could be ‘unpacked’ by close analysis of the statements. In this

way, the meaning of a statement was severed from any necessary reference to something

in the world. Thus, on this view, the meaning comprised the truth claims made in the

statement whether or not those truth claims were in fact true in the sense of

correspondence to the real world. For example, Russell illustrated his argument by

showing how reference to the (non-existent) king of France could be analysed into a

number of truth claims.76 Meaning existed, he claimed, without any need to refer to an

object in the world.

In the 1930’s the loosely affiliated members of the Vienna Circle of logical positivists

interpreted this quest for improving the objectivity of the language of science in various

ways. However, they shared the view that meaningful statements can only be made upon

‘objective’ foundations – comprising empirically verifiable propositions and analytic

truths (such as the logical truths of mathematics).77 According to this view, meaning

was found in the truth value of statements, and truths could be ‘analytic truths’ (because

the truth is contained within the proposition) or ‘synthetic truths proven empirically (a

posteriori truths)’.78 The title of Rudolf Carnap’s influential paper ‘The Elimination of

Metaphysics through Logical Analysis of Language’ captured this theme – the meaning

of language could be analysed by reducing words to their ultimate ‘observation

sentences’ which were truth evaluable – they could be at least theoretically open to

empirical confirmation.79 Ayer’s influential work disseminated this logico-empiricist

approach in Great Britain,80 arguing that the meaning of a statement is that which is

capable of derivation by analysing the statement or as its truth conditions which are

capable of empirical verification.81

While logical positivists were pursuing several approaches to eliminating metaphysics

from the language of science, inspired by the earlier work of Wittgenstein, from the

beginning of the 1930s Wittgenstein came to reject key elements of his earlier

philosophy.82 In his later work, Wittgenstein maintained that the meaning of words was

75 Bertrand Russell, ‘On Denoting’ (1905) 14 Mind 479. 76 Ibid. 77 The manifesto signed by some members of the Vienna Circle declared that ‘[c]larification of the

traditional philosophical problems leads us partly to unmask them as pseudo-problems and partly to

transform them into empirical problems and thereby to subject them to the judgment of empirical science.

The task of philosophical work lies in this clarification of problems and assertions, not in the propounding

of special “philosophical” pronouncements’. The Scientific Conception of the World: the Vienna Circle

(Reidel, 1973 [1929]) 8. 78 Immanuel Kant, Critique of Pure Reason (tr N Kemp Smith, St Martin’s Press, 1965 [1781]) 48. 79 Rudolph Carnap, ‘Uberwindung der Metaphysik durch der Logische Analyse der Sprache’ (‘The

Elimination of Metaphysics through the Logical Analysis of Language’) (1932) Erkenntnis II (reprinted in

A J Ayer, Logical Positivism (The Free Press, 1959)). 80 A J Ayer, Language, Truth and Logic (Victor Gollancz, 1956 [1936]). 81 Ayer, Language, Truth and Logic, above n 80, 35. 82 For careful comparison see P M S Hacker, ‘Wittgenstein on Grammar, Theses and Dogmatism’ (2012)

35(1) Philosophical Investigations 1.


19

conventionally determined and that language was an autonomous system, the words of

which bore no necessary relation to the things that they depict or name.83 However, this

does not mean that language use is arbitrary – language is a tool used to meet our

pragmatic ends84 and so we might develop conventions regarding the use of language

that are arbitrary in the sense that we might have developed different conventions in

pursuit of those pragmatic ends. But this does not make the rules of grammar arbitrary

in the strong sense of being incomprehensible because there is no way of knowing what

any of us have determined those rules to stand for. Nor are the rules of grammar

answerable to some correspondence with the real world or some ultimate pragmatic end

for which language is created. Thus, the study of the meaning of words requires the

study of the rules governing the proper use of components of natural languages.85

Wittgenstein suggested that the philosophical task was one of description, not analytical

discovery. By description, philosophical problems would be solved ‘through an insight

into the workings of our language … not by coming up with new discoveries, but by

assembling what we have long been familiar with’.86 Careful scrutiny of natural

language usage would reveal the nuances of the deeper layers of grammar beneath the

surface grammar of a particular text.87

4.3 The relevance of linguistic philosophy to analysis of Sir Frederick Jordan’s statement

This brief overview of the philosophy of language over the first decades of the twentieth

century indicates that different theories of meaning were being robustly assessed over

the period leading up to the time when Jordan CJ set out his definition of income. A

central theme of the literature of that period was the quest to refine the understanding

of language so that language would not be a barrier to truthful communication about the

world. This quest was grounded upon logical analysis or empirical verification of

statements made in natural languages:

1. one strand approached this subject upon the basis that logical analysis of the

semantic elements of truth claims would reveal the (possibly hidden) meaning

of statements; and

2. an alternative strand considered that examination of the conventional use of

language offered a therapeutic to philosophers who had turned their backs upon

the rationality of human thought reflected in common language usage.

Both approaches encountered fundamental difficulties in specifying the foundations of

truthful statements. A semantic theory of language confronted the challenge of

identifying the threshold at which a description was sufficiently detailed so as to identify

that which was signified. A conventional account of language meaning confronted the

challenge of specifying the threshold at which a usage was a convention and the basis

upon which that convention determined the statement’s meaning. Moreover, a

83 Wittgenstein, above n 73, [355]. 84 Wittgenstein, above n 73, [5] ‘clearly survey the purpose and functioning of the words’; see also [127],

[132]. 85 Wittgenstein, above n 73, [28]-[33]. 86 Wittgenstein, above n 73, [109]. 87 Wittgenstein, above n 73, [664].


20

conventional account must accept that ‘truth’ can only ever be what is right for us, which

is not consistent with the universality that we attribute to ‘truth’.88

This philosophical examination of the foundations of linguistic meaning is relevant to

the present examination of the dictum of Jordan CJ in several ways. First, linguistic

philosophy highlights the fact that the foundations of the meaning of linguistic elements

cannot be taken for granted. Second, the first proposition is reflected in the echoes of

the semantic and pragmatic threads, of both common understandings of the foundation

of linguistic meaning and also of linguistic philosophy, to be found in the dictum of

Jordan CJ, a matter taken up in section 5 of this article. Third, the examination of the

foundations of linguistic meaning, characteristic of linguistic philosophy, highlights the

matters that one might expect to be considered if reference to the dictum of Jordan CJ

is to reach beyond ritual incantation with judicial analysis of the elements of the dictum.

5. UNDERSTANDING SIR FREDERICK JORDAN’S STATEMENT THROUGH THE LENS OF

LINGUISTIC PHILOSOPHY

5.1 Science, and the science of language

In this intellectual context it would have been extremely difficult for lawyers to ignore

the underlying ‘scientific’ examination of language and its use.89 According to this

formalist90 approach to statutory construction, the meaning of legislation is an objective

fact revealed by appropriate methods of ‘statutory construction’. This objective

statutory meaning is then applied to transactions ‘objectively’ characterised according

to their respective legal forms. Indeed, in the taxation context, the proposition that

taxation could only be imposed by clear language had come to be adopted for more than

a century.91 If any reminder of this proposition were needed, decisions including Scott

v Cawsey92 and IRC v Duke of Westminster93 no doubt served to emphasise this narrative

of legal objectivism.

At the same time it would have been extremely difficult to ignore the fact that there was

no universal methodology that could be applied in discovering the objective linguistic

meaning in a manner that would sustain the legitimation of law upon the basis that law

comprised objective, determinate rules that could be applied with certainty. Judges had

accepted that the ordinary usage of words recorded in dictionaries did not

comprehensively capture the ‘natural or ordinary’ meaning of words.94 Further, by the

time of Sir Frederick’s decision in Scott, it had already come to be accepted that

88 Moore, above n 59, 297. 89 See, for example, Thomas C Grey, ‘Langdell’s Orthodoxy’ (1983) 45(1) University of Pittsburgh Law

Review 1. 90 For discussion of the concept of legal formalism, see Frederick Schauer, ‘Formalism’ (1988) 97(4) Yale

Law Journal 509; Judith Shklar, Legalism: Law, Morals and Political Trials (Harvard University Press,

2nd ed, 1986). For critical discussion of Shklar’s work, see Robin West, ‘Reconsidering Legalism’ (2003)

88(1) Minnesota Law Review 119. 91 For early expressions of the strict construction of penal legislation, including taxation law, see Ramsden

v Gibbs (1823) 1 B & C 319; 107 ER 119; Denn v Diamond (1825) 4 B & C 243, 245; 107 ER 1049, 1050. 92 (1907) 5 CLR 132, 154 per Isaacs J. 93 Inland Revenue Commissioners v Duke of Westminster [1936] AC 1, 24-25 per Lord Russell. 94 Girls’ Day Public School Trust v Ereaut [1931] AC 12, 34 per Lord MacMillan, observing that the editors

of the New Dictionary stated that the varieties of the sense of ‘public’ are such that there are many meanings

of the term. Also, it was accepted that the meaning of composite expressions could not be established by

merely compiling dictionary definitions of each part: Perpetual Trustee v FCT (1931) 45 CLR 224, 240 per

McTiernan J.


21

‘income’ had been ascribed different meanings in different contexts.95 In the early

twentieth century the ‘trust’ concept of income had been identified as a possible theory

of income for the purposes of Australian income taxation.96 Meanwhile, a commercial

or business understanding of income, influenced by the developing discipline of

accountancy, had been acknowledged to be an alternative theory of income that could

be applied in the context of income taxation.97 Just three years after the decision in Scott,

and during the era when logical positivists pursued their anti-metaphysical agenda,

Henry Simons contributed to stripping the economic concept of income of metaphysical

elements by focusing upon the apparent objectivity of ‘market value’ for the purpose of

identifying income as a tax base.98 Awash with these conflicting currents of thought

upon the nature of income, Greene MR captured the view of many with his suggestion

that the income/capital distinction seemed to turn on the spin of a coin.99

By the time of the decision in Scott the general statutory references to ‘income’

presented a considerable challenge to judges because of the tension between the ‘no

taxation without clear words’ maxim and the fact that ‘income’ did not clearly refer to

one finite meaning comprising necessary and sufficient criteria for its application to

particular circumstances.

In contexts other than the income tax this balancing of linguistic imprecision and, at

least the appearance of, legal certainty could be resolved by reliance upon principles of

interpretation that allowed reference to the statutory context100 or to the legislative

purpose.101 These principles allowed judges to construct, and quite possibly genuinely

believe in the truth of, a narrative of discovering the ‘right’ meaning out of a range of

possible meanings, while claiming that this right meaning had been the purpose of the

legislature all along (that only lawyers with superior skills, such as judges, could

discover). However, the purpose of taxing income was not considered by judges to offer

secure footing upon which to elaborate the meaning of income102 and, as Jordan CJ

noted,103 the statutory context of the reference to ‘income’ offered little guidance in

elaborating the meaning of income.

95 See the discussion of the different meanings of income in its general sense and for income taxation

purposes in Harding v Federal Commissioner of Taxation (1917) 23 CLR 119, 130 per Isaacs J (Gavan

Duffy and Rich JJ agreeing). 96 W Strachan, ‘The Differentiation of Capital and Income’ (1902) 18 Law Quarterly Review 274; W

Strachan, ‘Economic and Legal Differentiation of Capital and Income’ (1910) 26 Law Quarterly Review

40; W Strachan, ‘Capital and Income (Lifeowner and Remainderman)’ (1912) 28 Law Quarterly Review

175; W Strachan, ‘Capital and Income under the Income Tax Acts’ (1913) 29 Law Quarterly Review 163. 97 The Commissioner of Taxes (South Australia) v The Executor Trustee and Agency Company of South

Australia Limited (Carden’s Case) (1938) 63 CLR 108, 152 per Dixon J. 98 Henry Simons, Personal Income Taxation (University of Chicago Press, 1938) 42 and, more generally,

ch 2. 99 Inland Revenue Commissioners v British Salmson Aero Engines Ltd (1938) 22 TC 29, 43. 100 The Metropolitan Gas Co v The Federated Gas Employees’ Industrial Union (1924) 35 CLR 449, 452

(per Isaacs and Rich JJ). 101 Attorney General v Carlton Bank [1899] 2 QB 158, 164 (Lord Russell). 102 See, for example, the decision of Dixon J in Carden’s Case, The Commissioner of Taxes (South

Australia) v The Executor Trustee and Agency Company of South Australia Limited (Carden’s Case) (1938)

63 CLR 108, 152. 103 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 220.


22

5.2 Analytical approaches to ‘income’ reflected in Sir Frederick Jordan’s statement compared

to orthodox interpretation of non-technical statutory language

In this context we can begin to understand Sir Frederick Jordan’s reasoning behind

adoption of his forms/principles framework. This framework for interpreting the

definition of ordinary income is unusual when compared to the ‘ordinary grammatical

meaning’,104 ‘literal meaning’105 or ‘grammatical meaning’106 of the text commonly

referred to by judges when considering the general principles of statutory

construction.107 When these textual formulae are adopted it seems to be implicitly

assumed that the ordinary, natural or literal meaning is immediately apparent or readily

accessed, for example by reference to dictionaries.108

The first part of Sir Frederick Jordan’s statement appears to depart from this process of

identifying the meaning of non-technical language. In the same way that linguistic

philosophy accepted that rules governing the linguistic meaning could be revealed by

analysis, Sir Frederick indicated that language usage and concepts need to be analysed

in order to arrive at the meaning of natural language terms. Forms of receipt are

‘comprehended within’109 the word ‘income’, suggesting that forms of receipt will be

revealed by examining the inner logic of ‘income’. However, this inner logic can only

be revealed by examining ‘ordinary concepts and usages’. Sir Frederick seems to be

stating that identification of the forms of receipt comprehended within ‘income’ entails

both semantic analysis of the word and conventional analysis of the use of the word.

The preceding overview of linguistic philosophy indicates that this part of Sir

Frederick’s statement embodies a potentially irresolvable conflict between competing

semantic and pragmatic accounts of the foundation of linguistic meaning.

Sir Frederick’s reference to identifying ‘principles’ also presents challenges. It is not

clear whether the principles also are ‘comprehended within’ ‘income’ or whether they

are to be found by examining the social practice of using ‘income’. The phrase ‘ought

to be treated as income’ does not assist in clarifying the nature of the process

contemplated by Sir Frederick Jordan’s statement, because it could be referring to norms

evident on the surface of ordinary language usage or it could be referring to norms that

lie hidden beneath surface usage in circumstances where the usage does not necessarily

reflect those norms.110

Notwithstanding these difficulties, the first part of Sir Frederick’s statement has been

understood to offer a rigorous foundation for the elaboration of income by focusing

attention upon extracting forms and principles from ‘ordinary concepts and usages’.

104 Alcan (NT) Alumina Pty Ltd v Commissioner of Territory Revenue (NT) (2009) 239 CLR 27, 46 [45]. 105 Project Blue Sky Inc v Australia Broadcasting Authority (1998) 194 CLR 355, 384 [78]; Alcan (NT)

Alumina Pty Ltd v Commissioner of Territory Revenue (NT) (2009) 239 CLR 27, 46 [45]. 106 Project Blue Sky Inc v Australia Broadcasting Authority (1998) 194 CLR 355, 382 [72], 384 [78]. 107 While this ‘ordinary or natural’ etc meaning may not be adopted as the meaning of the legislative text

in the context of resolving a matter by adjudication, decisions such as Project Blue Sky proceed upon the

basis that such ‘ordinary or natural’ etc meaning should be adopted unless the context or purpose indicates

that a ‘legal’ meaning ought to be adopted. 108 This approach had long been adopted in Australia before the decision in Scott: Sydney Municipal Council

v Commonwealth (1904) 1 CLR 208, 241 per O’Connor J. Around the time of Scott, see Federal

Commissioner of Taxation v Riley (1935) 53 CLR 69, 80 per Starke J. 109 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 219 per Jordan CJ. 110 Because, for example, the general population may misunderstand the proper meaning of a term, by

reference to the ‘deep’ meaning of the term, and so misuse the term.


23

This interpretation appears to emphasise the analytical aspects of the statement while

downplaying the unresolved tension between semantic/conventional analysis and also

ignoring the vagueness of the source of the principles to be identified. By revealing the

hidden logic of ‘income’, Jordan CJ appeared to accept that the rules governing

classification as ‘income’ would be determinate at any particular point in time and that

these rules could underpin the classification of an amount as ‘income’ in circumstances

not previously encountered. Thus, the application of the concept ‘income’ could adjust

to changing social circumstances while remaining constant in its internal logic. This

balance between change and continuity therefore appeared to resolve the tension

between the ideal of legal certainty embodied in immutable law, reflected in the maxim

that taxation can only be imposed with certain words, and the apparent confusion of

competing concepts of income reflected in the contemporaneous literature.

The second part of Sir Frederick Jordan’s statement appears to be consistent with the

ordinary principle of adopting dictionary definitions of non-technical statutory

language. In the second sentence of the extract from Scott his Honour appears to

paraphrase his forms/principles approach by suggesting that we can determine whether

an amount is income simply by determining whether it is income ‘in ordinary parlance’.

Indeed, in the next paragraph in his judgment Sir Frederick observed that ‘according to

the ordinary meaning of “income,” as an English word, there can be no doubt that such

a receipt as that now in question is not income’.111 His Honour did not expressly apply

his forms/principles approach for the purpose of identifying what he took to be the

ordinary meaning of ‘income’, although this may be because Sir Frederick considered

such express analysis superfluous given his immediately preceding statement. However,

references to ‘ordinary parlance’ and to ‘the ordinary meaning of “income”’ could be

interpreted to contradict the analytical discovery of the deep meaning of income hidden

beneath the superficial veneer of ordinary usage. After all, ‘ordinary parlance’ is not

synonymous with ‘ordinary concepts and usages’ as, unlike the former, the latter is

broad enough to include written usages as well as parlance.

There is, then, a tension between the first and second parts of Sir Frederick’s statement.

If that statement were interpreted so that ‘income’ meant its meaning in ordinary

parlance as reflected in dictionary meanings, we would justifiably question why Sir

Frederick would have expressed such an orthodox proposition in such a convoluted

manner. Why refer to the forms/principles analysis at all? Likewise, the High Court

decisions that have adopted Sir Frederick’s statement might have more economically

adopted the orthodox principle regarding reference to dictionaries. At least in a formal

context such as a statute or a judicial decision, there is a conventional assumption that

all words are intended to have some effect, unless nonsense would arise by doing so.

Applying this interpretative assumption, all of the elements of Sir Frederick Jordan’s

statement should be recognised. However, in this case, it seems that a choice must be

made between incomparable approaches to identifying the meaning of income.

5.3 Does the forms/principles framework apply to other expressions that are not terms of art?

The forms/principles analysis is expressed in the first part of Sir Frederick’s statement

in such a manner as to suggest that this analysis ought to apply to all statutory terms that

are not terms of art and also to judicial terms that are not terms of art. Thus, ‘income’ is

not a term of art and its meaning is determined according to the forms/principles analysis

111 Scott v Commissioner of Taxation (1935) 35 SR (NSW) 215, 219 per Jordan CJ.


24

of ordinary concepts and usages and perhaps ‘ordinary parlance’. Is it also the case that

these expressions are not ‘terms of art’? If so, should the same forms/principles analysis

be undertaken in elaborating the meaning of these terms/phrases? Further, should the

same analysis be applied to those meanings, and so on in infinite regress unless at some

point this process will reveal ultimate, ‘atomic’ foundations that are true in virtue of

themselves? If there is an infinite regression, the foundation of meaning cannot be

found.112 However, Sir Frederick Jordan’s statement leaves this matter unresolved.

If it is possible for the meanings of components of Sir Frederick’s statement to be

identified without undertaking the same forms/principles analysis, upon what basis is

this determination made?

5.4 Why the forms and principles approach?

Elaboration and application of the two-stage forms/principles framework depends upon

the meanings attributed to ‘forms’ and ‘principles’ and also upon specification of the

source(s) from which those forms and principles are derived.

Sir Frederick’s statement clearly accepts that ‘income’ describes receipts that meet

particular formal descriptions. A lawyer reading the dictum, recorded in a legal

judgment, could reasonably be expected to read this reference to ‘forms of receipt’ to

be a reference to forms of receipt classified according to legal rules rather than amounts

being classified according to rules derived from non-legal sources. However, this legal

interpretation of ‘forms’ is undermined by two elements of the dictum of Jordan CJ. As

the forms of receipt are derived by analysis of ordinary concepts and usages, in this

context it would be reasonable to infer that his Honour did not have in mind the legal

form of a receipt. Moreover, this ‘non-legal’ interpretation of the reference to ‘forms’

is also supported by Sir Frederick’s observation that income is not a ‘term of art’. If

‘income’ were a term of art then it is more likely that legal forms would be considered

in characterising amounts, as distinct from applying non-legal rules when identifying

forms of receipt that are ‘income’.

In section 4.1 above it was noted that philosophers have refined different approaches to

classifying things such as income and that legal classifications of things such as income

are not necessarily consistent with these approaches. The reference to ‘forms’ might be

a passing acknowledgment of philosophical realism which, as we have seen, would

identify the essence of a particular amount by comparing it to the essence of a Platonic

Form or by comparing the amount against the formal Aristotelian essence or even by

assessing the amount against what Wittgenstein called a ‘logico-pictorial’ form.113 On

this view, ‘income’ might be equated to a genus that groups different species of receipt

according to their respective forms. Each form would have its own essence and hence

its own necessary and sufficient criteria. However, all forms would share the essence of

income.

However, his Honour’s reference to the secondary role of principles, also extracted from

ordinary concepts and usages, complicates this realist interpretation because it is

112 The prospect of a parallel infinite regression was raised by Quine in his critique of logical positivism:

W V O Quine, ‘Truth by Convention’ (1935) in W V O Quine, The Ways of Paradox, and Other Essays

(Harvard University Press, 1976) 77 and W V O Quine, ‘Two Dogmas of Empiricism’ (1951) 60(1)

Philosophical Review 20. 113 Ludwig Wittgenstein, Tractatus Logico-Philosophicus (Kegan Paul, 1922) 2.2.


25

difficult to see how ‘principles’ can be accommodated within a realist framework in

which things have an essence that is defined in terms of necessary and sufficient criteria.

If a receipt has a form of a kind that warrants its inclusion under the category ‘income’

because its essence corresponds with the essence of income, a realist would reject the

proposition that the particular amount cannot be income because ‘principles’ operate to

override the essential nature of the receipt. This suggests that Sir Frederick cannot have

meant to adopt some version of philosophical realism when referring to forms of receipt.

If Sir Frederick’s reference to ‘forms of receipt’ does not refer to legal formalism and

also does not refer to philosophical realism, what are the rules governing identification

of the relevant forms? And how do ‘principles’ interact with those rules? For example,

could principles affect the types of form comprehended within the concept of income?

These aspects of Sir Frederick’s statement remain unresolved.

Moreover, his Honour did not elaborate upon the reasons for adopting this dual

forms/principles inquiry. However, adopting the convention of interpreting statements

in their best possible light, it is reasonable to speculate that his Honour had in mind the

benefit of undertaking a preliminary cull of forms of receipt that could never satisfy the

concept of income described by the principles considered in the second stage of the

inquiry. For example, the relatively blunt instrument of assessing forms of receipts

could enable a judge to quickly dispose of many cases without recourse to the second

stage identification and application of principles. If a gains concept of income were to

be extracted from ordinary concepts and usages, benefiting from the public provision of

infrastructure could be considered to be a gain.114 However, exclusion of such benefits

upon the basis of their form (such as ‘absence of an ‘earning activity’115 or the absence

of a ‘receipt’116) could avoid the need for close consideration and application of the

second stage principles.

This forms/principles approach therefore has intuitive cognitive appeal because it

accords with two cognitive approaches that humans routinely adopt when categorising

things – application of necessary and sufficient criteria and also consideration of

concept theories when classifying more problematic cases.117 However, the cognitive

power of this two-stage analysis depends upon the specification of the forms and

principles and also the specification of the rules that prevent conflict between those

forms and principles. The absence of these definitional and operative rules in Sir

Frederick’s judgment and/or in judgments that have adopted Sir Frederick’s statement

means that it is not possible to take this analysis further here.

5.5 ‘Ordinary’

The references to ‘ordinary parlance’ and ‘ordinary concepts and usages’ maintain some

degree of continuity with established principles of interpretation that recognised the

‘ordinary or natural’ meaning of words.118 However, I have already noted that Jordan

114 See discussion of this point in Boris Bittker, ‘A “Comprehensive Tax Base” as a Goal of Income Tax

Reform’ (1967) 80(5) Harvard Law Review 925, 935. 115 Hayes v FCT (1956) 96 CLR 47, 54 per Fullagar J. 116 Payne v FCT (1996) 56 FCR 299. 117 These cognitive processes are discussed further below when dealing with concepts: see section 5.9

below. 118 Amalgamated Society of Engineers v Adelaide Steamship Co Ltd (1920) 28 CLR 129, 161-162 per

Higgins J.


26

CJ departed from that simplistic account of natural language meaning by suggesting that

analysis of ordinary concepts and usages and ordinary parlance would reveal the forms

of income receipt and principles to be applied in characterising receipts as income.

Although ‘income’ was to be elaborated by analysis and judgment, according to Jordan

CJ, the ordinariness of the language elements subjected to judicial analysis promoted a

narrative that depicted one concept of income that was routinely adopted in the

community and therefore was of the community and hence unremarkable. This narrative

promotes the view that judicial decisions regarding the income concept reflect the

univocal community’s understanding of the term rather than being imposed from above

by unrepresentative appointees to judicial office.119

The narrative of analysis of ordinary language usage also serves to explain why the

characterisation of amounts as income is not necessarily a straightforward application

of ordinary meaning and may require legal expertise to correctly characterise such

amounts. The narrative of analysis therefore promotes the legitimacy of ‘law as

scientific analysis of observable phenomena’ and also of the professional judgment

necessary to the operation of law.120 Judges, rather than ordinary folk, must be at the

centre of pronouncements upon the meaning of an ordinary term such as ‘income’.

Superficially, then, this narrative of linguistic analysis of ordinary language usage

answered several imperatives by portraying judicial practice as objective examination

of the community’s ‘ordinary’ meaning of language. However, given the central

importance of the concept of ‘ordinary’ to this narrative, surprisingly little attention has

been devoted to what is ‘ordinary’. Dictionary definitions reflect an ambivalence

regarding what is ordinary. One sense of the word is that it refers to the unexceptional

nature of the thing. Something may be rare but unexceptional – the last example of a

particular model of utilitarian passenger car, for example. ‘Ordinary’ might also refer

to the commonality of the occurrence and perhaps the distribution of the thing. William

Shakespeare’s poetry may be extraordinary by nature but so commonly available as to

be ordinary in the sense of availability/usage.

Turning to ‘ordinary concepts’, an extraordinary (by nature) concept might be so

frequently used that it is ordinary. Conversely, a concept that is unexceptional in terms

of the structure of the concept could be extraordinary because it is rarely used. Jordan

CJ did not elaborate upon the conditions governing what counts as ‘ordinary’, and the

dictionary definitions leave considerable leeways of choice.

In the absence of any judicial analysis of the concept of ordinariness itself for the

purposes of applying Sir Frederick’s statement, it is impossible to assess the truthfulness

of claims that a particular usage, concept or parlance is ‘ordinary’. Parsons alluded to

this epistemic problem when he suggested that members of the community might not

accept the concept of income attributed to them,121 as did Hill J in his first instance

119 See Harding v Federal Commissioner of Taxation (1917) 23 CLR 119, 130 per Isaacs J (Gavan Duffy

and Rich JJ agreeing), noting that the relevant sense of ‘income’ was one with which the community had

been familiar for over 100 years. 120 Eliot Friedson, Professionalism (University of Chicago Press, 2001) ch 1; noting Friedson’s caveats

about ‘ideal types’: 8-10. 121 Ross Parsons, ‘Income Tax - An Institution in Decay’ (1986) 3(3) Australian Tax Forum 233, 239.


27

decision in Stone.122 This absence of precision appears to have allowed some judges to

adopt a range of approaches to the concept of ordinariness which have underpinned

different approaches to the meaning of ‘income’. For example, a ‘business’ or

‘commercial’ concept of income could be adopted in characterising the nature of a

particular receipt123 while a gains concept of income can also be considered to be

‘ordinary’.124

5.6 Ordinary concepts and usages: a composite expression or a single object?

It is not clear whether ‘ordinary concepts and usages’ is one composite expression or

whether it refers to two sets of social phenomena: ‘ordinary concepts’ and ‘ordinary

usages’. An ordinary concept of ‘income’ could differ from an (ordinary?) usage of

‘income’, depending upon the meanings of ‘concepts’, ‘usages’, and ‘ordinary’. Is there

a difference between ordinary concepts and ordinary usage? If so, which one prevails

in arriving at the proper meaning of income? If ordinary concepts and ordinary usage

are different but lead us to the same meaning of income, why refer to them both?

5.7 Rules governing the identification of relevant ‘ordinary concepts and usages’

With respect to the epistemological aspect of Sir Frederick Jordan’s statement, his

Honour did not set out the basis upon which we ought to derive the relevant

forms/principles from the entire set of ‘ordinary concepts and usages’.

Confronted with the entire set of ordinary concepts and usages (ie, ‘star’, ‘fairy’, ‘apple’

are all concepts), how do we go about extracting the forms/principles specific to

‘income’? Without close specification of this inquiry, there is a risk that the

forms/principles which I identify will merely reflect personal predispositions regarding

the nature of income rather than having an objective logical or empirical foundation. To

illustrate this point, it is fair to suppose that there are ordinary concepts and usages with

respect to ‘apples’. It is also fair to suppose that most would accept that ordinary

concepts and usages with respect to apples have very little or nothing to do with ordinary

concepts and usages regarding the forms/principles of ‘income’. But upon what basis

can I be sure that concepts and usages regarding apples have little or nothing to do with

income other than because I have already formulated a concept of income that

incorporates an element that allows me to exclude apples, or, a concept of apples that

excludes income?

122 Stone v FCT [2002] FCA 1492, 2002 ATC 5085, 5094 [57], noting that some observers may be surprised

that ordinary concepts are so complex, but continuing by noting the ambiguity commonly encountered in

natural languages. 123 The Commissioner of Taxes (South Australia) v The Executor Trustee and Agency Company of South

Australia Limited (Carden’s Case) (1938) 63 CLR 108, 152. The recognition of the ordinariness of the

commercial and accounting concept of ‘profits’ in the context of the United Kingdom income tax legislation

had been accepted in Gresham Life Assurance Society v Styles [1892] AC 309; 3 TC 185, 189 per Halsbury

LC (‘in a sense which no commercial man would misunderstand’); 3 TC 185, 191 per Lord Herschell (‘as

a matter of business’). 124 See, for example, Commercial and General Acceptance Ltd v FCT (1977) 137 CLR 373, 382 per Mason

J. Note also the adoption as ‘ordinary’ of a gain concept of income by Pitney J in Eisner v Macomber 252

US 189, 206-207 (1920); see nn 54-55 above. For discussion of this aspect of the decision in Montgomery

see Domenic Carbone, ‘An Extraordinary Concept of Ordinary Income? The Significance of FCT v

Montgomery on What is Income According to Ordinary Concepts’ (2010) 20 Revenue Law Journal 1, 21-

24.


28

5.8 ‘Parlance’

I have already noted that the reference to ‘ordinary parlance’ could restrict the scope of

the survey of ordinary concepts and usages to those concepts and usages found in

‘ordinary conversation’. This privileging of the spoken word, upon the basis that oral

utterances offered an immediate reflection of the speaker’s meaning that is lost when

words are committed to writing, was subjected to critical analysis by Derrida.125 Without

revisiting that critique, for present purposes it is sufficient to note that the reference to

‘parlance’ indicates that the analytical narrative embodied in Sir Frederick’s statement

could be taken to suggest that judges focus upon analysing the one true meaning of

income revealed in ordinary parlance, by contrast to analysis of ‘degenerate’ usages of

‘income’ such as those found in written form. If this interpretation is correct, there

would be a contradiction in reflecting ordinary parlance in written judgments and also

in analysing written usages of ‘income’ in documents such as written case decisions.

5.9 Theories of Concepts

Subsection 6-5(1) specifically refers to ‘ordinary concepts’ and therefore could be taken

to adopt the same phrase from Sir Frederick Jordan’s statement in Scott. Given the

importance of the concept of a concept to identifying the statutory meaning of ‘ordinary

income’, it is surprising that the concept of a concept has not been subjected to close

analysis in the case law regarding the meaning of income.

The purpose of this section is to provide an overview of different concepts of concepts

in order to take Australian tax jurisprudence some way down the path of addressing that

lacuna. In elaborating upon different theories of concepts, I am not attempting to

develop an argument for a preferred concept of a concept. Rather, the first purpose of

this section is to establish the proposition that there is a multitude of concepts of

concepts available and that these concepts are ‘ordinary’ in terms of their general

recognition and/or in terms of their use.126 If there are different ordinary concepts of

concepts, and different ordinary concepts of concepts such as ‘income’ and ‘capital’,

the difficulty of applying the statutory definition in subsection 6-5(1) is compounded.

This difficulty is only exacerbated by the fact that the choice between concepts of

concepts is not guided by statutory, judicial or non-legal rule.127 If these premises are

accepted, then the meaning of ‘ordinary income’ is inherently unstable. Therefore legal

pluralism or legal scepticism could more accurately describe the operation of the

statutory definition of ‘ordinary income’ than legal formalism.128

125 Jacques Derrida, ‘Signature Event Context’ in Limited Inc (Northwestern University Press, 1988) ch 1. 126 I hope that the accompanying references are useful for those wishing to engage with the literature upon

concepts of concepts. 127 Typically, judges accept that any statutory term has an ordinary, natural, literal or grammatical meaning.

The basis upon which such a meaning is identified is not specified. It seems that it is simply assumed that

one such meaning exists. One anonymous reviewer suggested that my approach to ‘concepts’ was

misconceived, upon the basis that the statutory and judicial references to ‘concepts’ was to the ordinary or

natural meaning of the word: The author makes some interesting points in part 9, ‘Theories of Concepts’, but, if I interpret her

correctly, in doing so seems to assume that Jordan CJ intended to use the term ‘concepts’ with

one of a number of technical meanings, when, as explained, he appears more probably to have

intended simply to expound the ordinary meaning rule of standard statutory interpretation.

The point of this section, as I say in this paragraph, is that there is no one ordinary meaning of ‘concept’. 128 Compare Evans’ suggestion that a ‘theory theory’ of concepts can both recognise competing theories of

constitutional meanings while simultaneously providing a stable foundation for constitutional


29

5.9.1 Dictionary definitions

The Shorter Oxford English Dictionary states that a ‘concept’ is an idea of a class. The

Shorter Oxford English Dictionary also indicates two relevant senses in which

‘ordinary’ is commonly used: ‘ordinary’ according to the common usage of the thing

and ‘ordinary’ according to the intrinsic nature of the thing. Putting these elements

together, it is plausible that an ‘ordinary concept’ is any idea of a class of objects that is

widely found or widely attested in language usage (rather than being a specialised, or

what might be called an ‘artificial’ language, such as that of symbolic logic). I have also

noted that the ITAA 1997 has not stipulated the meaning of the phrase ‘ordinary

concepts’ and nor have relevant case decisions analysed the meaning of the phrase.

5.9.2 Concepts of concepts in the literature of philosophy and psychology

For those with even limited knowledge of the literature upon the subject of concepts,

the propositions that there are multiple theories of concepts, and that to date there is no

one ‘universally right’ concept of a concept has been identified, will be uncontroversial.

However, the apparent judicial reluctance to undertake an analysis of concepts of

concepts makes it necessary for me to sketch some of the main concepts of concepts.

Laurence and Margolis129 identify five broad categories of concepts of concepts,

conceding that this taxonomy does not do justice to the many nuances identified in the

literature upon this subject. Those categories are:

1. classical;

2. prototype;

3. exemplar;

4. ‘theory theories’; and

5. various combinations of the preceding categories.

In the following paragraphs I will sketch these concepts and illustrate their respective

relevance to inquiries into the meaning of ‘income’.

5.9.3 Classical theories of concepts

Classical theories of concepts characterise concepts as definitions that denote the

necessary and sufficient criteria for a concept to apply in a particular case. According

to this view, an individual possesses a concept when they grasp the rules by which a

member of a set is identified – when they know the necessary and sufficient conditions

for membership of the relevant class.130 This definitional account of concepts is not only

interpretation. Evans does not explain how this stability can arise, given his apparent acceptance that there

is no basis upon which the merits of competing theories can be assessed and the one stable meaning

identified: Simon Evans, ‘The Meaning of Constitutional Terms: Essential Features, Family Resemblance

and Theory-Based Approaches’ (2006) 29(3) University of New South Wales Law Journal 207. 129 Stephen Laurence and Eric Margolis, ‘Concepts and Cognitive Science’ in Stephen Laurence and Eric

Margolis (eds), Concepts, (MIT Press, 1999) ch 1. 130 Joseph Raz adopts different definitions of concepts at different points in his work. However, at one point

he appears to adopt a classical definition of concepts, stating that ‘I will follow them in equating complete

mastery of a concept with knowledge and understanding of all the necessary features of the objects to which

it applies. Thus, complete mastery of the concept of a table consists in knowledge and understanding of all

the essential properties of tables, and so on’. Between Authority and Interpretation, above n 58, 21.


30

applicable for those who subscribe to philosophical realism,131 because the definition

may identify the necessary and sufficient criteria that must exist in reality (the real

essence of the thing) or they may identify the necessary and sufficient criteria that must

be satisfied by convention.132

The classical theory of concepts resonates with legal rule formalism.133 According to

rule formalism, the meaning of a legal concept (such as income) is identified having

regard to its set of definitional rules and this set of definitional rules is then ‘laid on top’

of the ‘fact pattern’ comprising our sensory perceptions of the particular instance under

examination. One can then mechanically determine whether or not the ‘fact pattern’

corresponds to the definitional rules of the concept.

Although classical theories of concepts were the predominant paradigm for concepts

until the latter half of the twentieth century,134 they have been subjected to considerable

criticism in contemporary philosophy and cognitive psychology upon several

grounds.135 One limitation is that it is extremely difficult to identify a concept that

satisfies the classical focus upon definitional rules that comprehensively specify

necessary and sufficient conditions for their application.136 A second difficulty is that

the classical theory implies that all members of a class would be identified with more

or less equivalent speed because of the binary nature of determining whether each of

the necessary and sufficient conditions were respectively satisfied. However,

experimental data indicates that this is not the case because individuals exhibit

‘typicality effects’ in processing the application of a concept – ‘simple’ cases are more

readily identified to fall within a concept than ‘difficult’ cases.137

One explanation for these typicality effects is that the definitional rules comprising

concepts may be fuzzy. With respect to the concept of income, the High Court appears

to have accepted that this is the case. In Anstis, the majority’s joint judgment appeared

to acknowledge that a criterial approach to identifying the concept of income was not

discernible in the income tax case law:

There is a difficulty in making good absolute propositions in this field. In

Federal Commissioner of Taxation v Montgomery, Gaudron, Gummow, Kirby

and Hayne JJ recognised that ‘income is often (but not always) a product of

131 See the discussion in section 4.1.1 above. 132 A point noted, for example, by Edouard Machery, Doing Without Concepts (Oxford University Press,

2009) 78. Thus, the classical theory of concepts is not describing a universal thing of the kind of a Platonic

Form. 133 See, for example, the discussion of ‘serious rules’ in Larry Alexander, ‘“With me it’s all er nuthin”:

Formalism in Law and Morality’ (1999) 66(3) University of Chicago Law Review 530, 539; Schauer, ‘Legal

Formality’, above n 90. 134 Chris Swoyer, ‘Conceptualism’ in P F Strawson and Arindam Chakrabarti (eds), Universals, Concepts

and Qualities: New Essays on the Meaning of Predicates (Routledge, 2017) 127, 141; Machery, above n

132, 77-78. 135 R Mark Sainsbury, ‘Concepts without Boundaries’ in Rosanna Keefe and Peter Smith (eds), Vagueness:

A Reader (MIT Press, 1996) 186; J Fodor, M Garrett, E Walker and C Parkes, ‘Against Definitions’ (1980)

8(3) Cognition 263. 136 Wittgenstein, for example, argued that the concept ‘game’ could not be defined in a manner that satisfied

the classical idea of concepts: Wittgenstein, above n 73, [65]-[78]. 137 Laurence and Margolis, above n 129, 24-26. Margolis and Laurence critically assess the limitations of

the classical account of concepts that have been identified in the literature.


31

exploitation of capital; income is often (but not always) recurrent or periodical;

receipts from carrying on a business are mostly (but not always) income’.138

By contrast, the High Court in Montgomery appeared to adopt a criterial approach in

identifying the concept of the class of all inquiries into the meaning of income by

specifying the essential nature of such inquiries.139 However, the vagaries of Sir

Frederick’s statement, already noted earlier in this article, indicate that the elements

identified in his Honour’s statement are insufficient to constitute necessary and

sufficient conditions for identifying inquiries into the income concept which would

satisfy the requirements of Montgomery.

5.9.4 Prototype theories

Prototype theories of concepts emerged in the 1970s as research was undertaken

indicating that individuals did not necessarily apply the classical theory of concepts in

practice.140 Prototype theories relax the strictness of necessary and sufficient conditions

by suggesting that concepts include a body of knowledge regarding the statistical

ordering of properties most likely to be found in members of a class.141 This probabilistic

understanding of concepts does not necessarily, and typically does not, maintain that

concepts exclusively comprise this statistical information.142 However, a prototype

concept allows for quick classification of many cases which exhibit properties typically

found in the relevant concept and therefore the prototype theory offers an explanation

of cognitive studies which suggest that the classical concept of concepts is not always

applied in practice.143

Machery notes that prototypical properties can be featural or dimensional.144 A featural

property is one that a member of a class either has or does not have. Income is not

capital, for example. A dimensional property is one that a member of a class can have

to varying degree. Parsons’ seminal identification of indicia of income implicitly

recognised a prototype concept of income by identifying the features of income

commonly found in the set of income amounts without specifying that these indicia

were necessary.145 Regularity of receipt, for example, often exists when an amount is

income but regularity of receipt is neither determinative146 of the ‘characterisation as

income’ question nor is it essential.147

138 Federal Commissioner of Taxation v Anstis (2010) 241 CLR 443, 452 [19]. 139 Federal Commissioner of Taxation v Montgomery (1999) 198 CLR 639, 661 [64]. 140 Prototype theory grew out of Wittgenstein’s reservations regarding the usefulness of a classical theory

of concepts: Wittgenstein, above n 73, [65]-[78]. 141 This approach arose because of the difficulties encountered in framing comprehensive descriptions of

the objects of language. To overcome this difficulty with respect to Russell’s theory of descriptions, Searle

anticipated the prototype model by suggesting that a proper name (ie, ‘Aristotle’) did not necessarily require

a comprehensive description of all necessary elements that make up ‘Aristotle’: John Searle, ‘Proper

Names’ (1958) 67 Mind 166. 142 Machery, above n 132, 85. 143 Laurence and Margolis, above n 129, 29-30. 144 Machery, above n 132, 84. 145 Ross Parsons, Income Taxation in Australia (Law Book Co, 1985) ch 2. 146 Federal Commissioner of Taxation v Anstis (2010) 241 CLR 443, although query whether the import of

this decision is that regularity may support an inference of dependence or reliance and that the two factors

combined will be sufficient to characterise an amount as income. 147 Commissioner of Taxation v Myer Emporium Ltd (1987) 163 CLR 199, 210.


32

One source of difficulty with concept prototypes is that they might include statistical

information about dimensional features, such as the mean and standard deviation from

the mean. But such information is fuzzy, as atypical members of a class exist. The

extract from Anstis in the preceding section of this article148 illustrates why this

statistical ordering of dimensional features of the concept ‘income’ could be applied for

Australian income tax purposes. However, such a statistical ordering of dimensional

features is problematic, because no one property can have a high indexical rating owing

to the fact that there are cases in which each property, respectively, did not exist. A one-

off receipt was income,149 a non-receipt was income because there was a profit150 and

an amount that did not fall within one of the three generally recognised categories of

receipt was income,151 and so on.

A second source of difficulty with concept prototypes is that the rules governing

specification of the featural and dimensional properties are ill-defined. While there is

evidence to suggest that individuals use a prototypical approach to concepts in day-to-

day living,152 as it seems the High Court did in Anstis,153 a key difficulty confronting

prototype theories is specifying the rules by which we identify the statistically relevant

features.154 Birds, for example, have many features that are either ignored or

downplayed when applying the concept ‘bird’ (ie, digestive system, circulatory system,

possession of a brain, eyes). Upon what basis do we focus upon just some of these

features in specifying the prototypical properties?

5.9.5 Exemplar theories

Exemplar theories conceive of concepts as a body of information regarding examples

of a concept encountered in the past. Reasoning by analogy, this information regarding

the properties of prior examples is compared to the thing under scrutiny and a

classification decision is made. Exemplar theories of concepts account for the

categorisation of a thing as being a member of a concept class upon the basis that the

thing closely resembles a known member of the class. When crossing difficult

epistemological terrain, reasoning by (dis)analogy to past cases makes sense and is

commonly found in case law dealing with the meaning of income.

Consider the case of a thing that closely matches a known member of a class but is only

‘moderately similar’155 to other members of the class. This thing is more likely to be

classified as a member of the class by comparison to another thing that is only

moderately similar to most known members of the class. This aspect of exemplar

theories derives from the fact that they typically recognise that the properties taken into

account are dependent, in that application of a concept is a non-linear function of the

properties of the subject that match the exemplar. Machery observes that the non-

linearity of the exemplar function distinguishes exemplars from prototypes because

prototypes typically apply a linear function.156

148 See text accompanying n 138 above. 149 For example, Commissioner of Taxation v Myer Emporium Ltd (1987) 163 CLR 199. 150 Warner Music Pty Ltd v FCT (1996) 70 FCR 197. 151 Federal Commissioner of Taxation v Anstis (2010) 241 CLR 443. 152 Laurence and Margolis, above n 129, 84. 153 Federal Commissioner of Taxation v Anstis (2010) 241 CLR 443. 154 Machery, above n 132. 155 For discussion of this aspect of exemplar theories, see Machery, above n 132, 98. 156 Machery, above n 132, 98.


33

So when considering whether an amount is ‘ordinary income’, an exemplar theory

suggests that if the receipt matches my exemplars of income in several respects

(beneficially received, etc) but does not match other properties of my exemplars (ie, the

receipt is irregular) then I apply a function (multiplicative, for example) in weighing

whether the receipt is income. In the absence of express consideration of this function,

it is possible for different judges and others to apply different functions in resolving a

particular case, a process loosely described as determining the weight to be given to

particular facts.

The unspecified application of this function underpinned the reservations regarding

analogical reasoning – application of an exemplar concept of concepts – expressed by

the majority in Montgomery. There the majority observed that counsel in their

submissions had argued upon the basis of what they considered to be analogous cases,

an approach that the majority justices stated missed the import of the definition of

income.157

5.9.6 ‘Theory theories’

One of the key difficulties with both prototype theories and exemplar theories is that

they do not currently specify the basis upon which prototypes or exemplars are

identified. ‘Theory theories’ of concepts seek to address this shortcoming by suggesting

that concepts are theories or else are part of a broader theoretical/conceptual framework.

Reflecting this ‘theory theory’ of concepts, both Parsons158 and Prebble159 have

criticised the concept of income adopted for income tax purposes upon the basis that it

does not reflect any theory of the nature of income.

In broad terms, the proposition that concepts are theories is taken to mean that each

concept comprises information that explains the membership of a category. Margolis

and Laurence suggest that this idea of theories as explanations is tied to the idea that

scientific theories explain phenomena and that this idea of theories is supported by the

many instances of essentialist thinking that psychologists observe.160 Essentialist

conceptualisation entails identification of the essence of a thing – a three-legged dog

may have lost its tail, no longer bark and suffer from a condition that means that it has

lost its fur, but it will still be a dog because it satisfies the hidden essence that causes it

to be a dog.161 Thus, a ‘theory theory’ of concepts does not countenance the checklist of

observable properties envisaged by classical theories. One key benefit of ‘theory theory’

is that it appears to explain the development of concepts as children mature into

adulthood,162 and so it is understandable that developmental psychologists have been

most attracted to the ‘theory theory’ of concepts.

Under the second type of ‘theory theory’, concepts are parts of a broad theoretical

framework within which the mind organises concepts into domains. Machery suggests

that ‘domains are sets of entities, including categories, properties, and processes, that

157 See extract accompanying n 35 above. 158 Parsons, ‘Income Tax - An Institution in Decay’, above n 121. 159 Prebble, above n 62 and Prebble, above n 63. 160 Laurence and Margolis, above n 129, 45-6. 161 S Gelman and H Wellman, ‘Insides and essences: Early understandings of the non-obvious’ (1991) 38

Cognition 213, cited by Machery, above n 132, 103. 162 Laurence and Margolis, above n 129, 46-47.


34

are treated similarly by the mind’.163 Thus, for example, artificial and natural domains

may be separated upon the principle of differentiating those things that have been

created intentionally from those that arose without deliberate action. Accidental

inventions may pose a challenge for this neat categorisation, but the point is that the

‘theory theory’ envisages a metatheoretical structure within which concepts (possibly

comprising mini-theories) will be organised.

In the context of the income tax, a ‘theory theory’ of income could be grounded upon

some concept of justice. Benefit theory164 and gains theory165 are alternate options that

might shape our thinking about what falls within the class of ‘income’, either generally

or more specifically in the context of taxation. Alternatively, the subjective intention of

the taxpayer (or objective/subjective intention) might offer a similar foundation for

explaining categorisation of particular amounts as income.166 If there is a change to a

part of the underlying metatheoretical structure, it is reasonable to expect that this

flapping of a butterfly’s wings might generate ripples throughout the entire conceptual

structure as theories of different concepts are adjusted to take account of the changes as

they occur. Thus, if the class of things in the world that is grouped under a particular

concept changes, then this could affect the scope of other concepts. For example, the

concept of ‘capital’ might change, prompting a reconsideration of the concept of

income. Likewise, the concept of ‘business activity’ might change, once again

prompting a reconsideration of ‘income’ and ‘capital’.167 This approach clearly has

implications for the durability of particular concept definitions, as they are liable to

change at any moment as a result of change to any other conceptual definition.

5.9.7 Combinations of conceptual theories

So far we have been examining individual theories of concepts upon the basis that a

person applies the one theory of concepts with respect to all concepts and with respect

to all aspects of those concepts. We have noted that depicting broad concepts of

concepts in this way is undertaken with crude brushstrokes that obscure the nuances of

individual theories and that individual theories may contemplate that concepts are a

complex assemblage of different concepts of concepts. Some theories of concepts take

this heterogeneity further in accepting that individuals apply combinations of concepts

of concepts in higher level thought.

These hybrid theories suggest that individuals use different conceptual theories with

respect to different aspects of a concept, but nevertheless arrive at a fixed concept

comprising different conceptual approaches.168 These heterogeneous theories also

accept that individuals apply different concepts of concepts, but argue that individuals

may have different concepts for the same thing and that the application of the most

appropriate concept will be determined having regard to the context in which the

conceptual thinking occurs.

163 Machery, above n 132, 103. 164 Graeme Cooper, ‘The Benefit Theory of Taxation’ (1994) 11(4) Australian Tax Forum 397. 165 Simons, above n 98. 166 For discussion of intention of the artefact maker, see Paul Bloom, ‘Intention, history, and artefact

concepts’ (1996) 60(1) Cognition 1, cited in Machery, above n 132, 103. 167 See discussion of this point in Commissioner of Taxation v Stone (2005) 222 CLR 289, 295-297. 168 Machery, above n 132, ch 3.


35

5.9.8 Is there an ordinary concept of a concept?

The preceding overview of different concepts of concepts indicates that there is

considerable difficulty in identifying those concepts which are ordinary, in the sense of

widely used or attested.

If ‘widely used or attested’ is understood to imply that something is preferred, then it is

possible that the set identified by the label ‘ordinary concept’ is an empty set because

no concept of a concept is preferred. Rather, it is possible that different ideas of concepts

are applied by different people and in different contexts. However, I suggest that it is

appropriate to adopt the interpretative presumption against meaningless statutory words

by proceeding upon the basis that ‘widely used’ does not require that a particular

concept of a concept is preferred. Rather, widely used can be understood to mean

‘commonly found in usage’. Thus, given that at least some of the different concepts of

concepts described here are commonly accepted to be found in common usage, the

legislative reference to ‘ordinary concepts’ is not an empty set.

However, proceeding on this basis does not resolve the difficulty of identifying ordinary

concepts because the ordinary concept of a concept could include all manner of theories

of concepts. If so, a judge must extract a concept of income after considering a multitude

of possibly incomparable, ordinary concepts of income, and that choice is not guided

by any principle expressed in the legislation or the case law.

6. CONCLUSION

In Montgomery a majority of the High Court stated that at least some elements of the

statement of Sir Frederick Jordan in Scott identified the essential nature of the inquiry

into the meaning of income and that this statement was not a matter of ritual incantation.

In doing so, the High Court juxtaposed ritual incantation with analysis, implying that

statutory meaning is an object that can be discovered by analysis of the relevant statutory

text in accordance with orthodox principles of statutory interpretation.

Despite this legalist depiction of the objective discovery of law, the meaning of the

subsection 6-5(1) definition of ‘ordinary income’ remains shrouded in a Dickensian

fog169 that several authoritative High Court decisions upon the subject have not

dispersed. Indeed, with respect, those High Court decisions appear to have thickened

the fog by adopting various inconsistent statements of the elements of Sir Frederick

Jordan’s statement. Moreover, those decisions have not analysed the elements of that

statement. Given this significant lacuna in the High Court’s elaboration of the concept

of ordinary income, this article analysed Sir Frederick Jordan’s statement by locating it

in the context of the vibrant linguistic philosophy of the early twentieth century. Viewed

through the lens of this contemporaneous linguistic philosophy, analysis of Sir

Frederick’s statement suggests that the meaning and interaction of the elements of that

statement are ill-defined. Further, it was noted that some of those elements may be

contradictory, giving rise to the possibility that Sir Frederick Jordan’s statement

incorporates irresolvable contradictions.

These conclusions regarding the absence of analysis undertaken by the High Court and

the difficulties that would arise were such analysis were to be undertaken, are significant

for several reasons. They are significant for the doctrinal approach to the concept of

169 Charles Dickens, Bleak House (Penguin, 1971 [1853]) ch 1.


36

ordinary income, for our self-understanding as we reflect upon the nature of ‘law’170

and also to the sociology of law.

From the perspective of doctrinal elaboration of the concept of income, the analysis of

Sir Frederick Jordan’s statement indicates that the statement does not offer a secure

foundation for identifying ‘income’. This indicates that a different approach to the

statutory definition of ‘ordinary income’ must be found. However, the plurality of

concepts of concepts poses considerable challenges if the statutory definition of

ordinary income is to be interpreted in line with orthodox principles of statutory

construction.

If there is no uniformly adopted foundation upon which amounts are characterised under

the income/capital dichotomy, there is the possibility that law is inherently pluralist as

Davies suggests.171 Taking this inherent heterogeneity of law to an extreme, it is possible

that this instance of what is commonly described as ‘black letter law’ in fact is one in

which philosophical scepticism might offer a more appropriate account172 than theories

of law framed upon determinate legal meaning173 or limited ‘judicial discretion’.174

Moreover, as ‘judicial activism’175 is necessarily part of the practice of Australian

taxation law because of the absence of analysis described in this article, and also because

of the intrinsic pluralism that would be exposed if analysis were undertaken,

sociological examination of law’s legitimacy in this domain should be re-examined. If

the legitimacy of judicial practice in this domain cannot plausibly be grounded upon the

objective discovery of statutory meaning, then the existence and foundations of that

legitimacy need to be reassessed.

170 Raz, Between Authority and Interpretation, above n 58, 31. 171 Margaret Davies, ‘The Ethos of Pluralism’ (2005) 27(1) Sydney Law Review 87. 172 Prebble, ‘Why is Tax Law Incomprehensible?’, above n 63. 173 For example, Ronald Dworkin, above, n 58. 174 H L A Hart, The Concept of Law (Oxford University Press, 2nd ed, 1994) ch 7. 175 Dyson Heydon, ‘Judicial Activism and the Death of the Rule of Law’ (2003) Quadrant, January-

February, 9; for critical review of this article, see Allan C Hutchinson, ‘Heydon’ Seek: Looking for Law in

All the Wrong Places’ (2003) 29(1) Monash University Law Review 85.

Documents

Interpreting the Australian income tax definition of ... · This article takes the High Court’s rejection of ritual incantation seriously by critically analysing the elements of